[r-cran-erm] 18/33: Import Upstream version 0.14-0
Andreas Tille
tille at debian.org
Mon Dec 12 11:19:34 UTC 2016
This is an automated email from the git hooks/post-receive script.
tille pushed a commit to branch master
in repository r-cran-erm.
commit 8e387bd38d8471a6ef6b9d1f5bdad295d0be3e6c
Author: Andreas Tille <tille at debian.org>
Date: Mon Dec 12 11:20:03 2016 +0100
Import Upstream version 0.14-0
---
COPYING | 0
COPYRIGHTS | 0
DESCRIPTION | 52 +-
MD5 | 183 ++
NAMESPACE | 12 +
NEWS | 24 +-
R/IC.default.R | 0
R/IC.ppar.r | 0
R/IC.r | 0
R/LLRA.R | 16 +
R/LLTM.R | 0
R/LPCM.R | 0
R/LRSM.R | 0
R/LRtest.R | 0
R/LRtest.Rm.R | 0
R/MLoef.R | 0
R/NPtest.R | 55 +-
R/PCM.R | 0
R/RM.R | 0
R/ROCR_aux.R | 0
R/RSM.R | 0
R/Waldtest.R | 0
R/Waldtest.Rm.R | 0
R/anova.llra.R | 31 +
R/build_W.R | 21 +
R/checkdata.R | 0
R/cmlprep.R | 0
R/coef.eRm.R | 0
R/coef.ppar.R | 0
R/collapse_W.R | 10 +
R/confint.eRm.r | 0
R/confint.ppar.r | 0
R/confint.threshold.r | 0
R/cwdeviance.r | 0
R/datcheck.LRtest.r | 0
R/datcheck.R | 0
R/datprep_LLTM.R | 0
R/datprep_LPCM.R | 0
R/datprep_LRSM.R | 0
R/datprep_PCM.R | 0
R/datprep_RM.R | 0
R/datprep_RSM.R | 0
R/fitcml.R | 0
R/invalid.R | 0
R/itemfit.R | 0
R/itemfit.ppar.R | 0
R/labeling.internal.r | 0
R/likLR.R | 0
R/llra.datprep.R | 43 +
R/llra.internals.R | 107 +
R/logLik.eRm.r | 0
R/logLik.ppar.r | 0
R/model.matrix.eRm.R | 0
R/performance.R | 0
R/performance_measures.R | 0
R/performance_plots.R | 0
R/person.parameter.R | 0
R/person.parameter.eRm.R | 0
R/personfit.R | 0
R/personfit.ppar.R | 0
R/pifit.internal.r | 0
R/plist.internal.R | 0
R/plot.ppar.r | 0
R/plotCI.R | 0
R/plotDIF.R | 287 +-
R/plotGOF.LR.R | 42 +-
R/plotGOF.R | 0
R/plotGR.R | 54 +
R/plotICC.R | 0
R/plotICC.Rm.R | 0
R/plotPImap.R | 0
R/plotPWmap.R | 0
R/plotTR.R | 31 +
R/plotjointICC.R | 0
R/plotjointICC.dRm.R | 0
R/pmat.R | 0
R/pmat.default.R | 0
R/pmat.ppar.R | 0
R/prediction.R | 0
R/print.ICr.r | 0
R/print.LR.R | 0
R/print.MLoef.r | 2 +-
R/print.eRm.R | 0
R/print.ifit.R | 0
R/print.llra.R | 10 +
R/print.logLik.eRm.R | 0
R/print.logLik.ppar.r | 0
R/print.pfit.R | 0
R/print.ppar.R | 0
R/print.resid.R | 0
R/print.step.r | 0
R/print.summary.llra.R | 20 +
R/print.threshold.r | 0
R/print.wald.R | 0
R/residuals.ppar.R | 0
R/sim.2pl.R | 0
R/sim.locdep.R | 0
R/sim.rasch.R | 0
R/sim.xdim.R | 0
R/stepwiseIt.R | 0
R/stepwiseIt.eRm.R | 0
R/summary.LR.r | 0
R/summary.MLoef.r | 0
R/summary.eRm.R | 0
R/summary.llra.R | 17 +
R/summary.ppar.R | 0
R/summary.threshold.r | 0
R/thresholds.eRm.r | 0
R/thresholds.r | 0
R/vcov.eRm.R | 0
R/zzz.R | 8 +-
data/llraDat1.rda | Bin 0 -> 2569 bytes
data/llraDat2.rda | Bin 0 -> 1187 bytes
data/llradat3.rda | Bin 0 -> 324 bytes
data/lltmdat1.rda | Bin 955 -> 642 bytes
data/lltmdat2.rda | Bin 125 -> 121 bytes
data/lpcmdat.rda | Bin 213 -> 211 bytes
data/lrsmdat.rda | Bin 317 -> 315 bytes
data/pcmdat.rda | Bin 210 -> 208 bytes
data/pcmdat2.rda | Bin 710 -> 478 bytes
data/raschdat1.rda | Bin 957 -> 641 bytes
data/raschdat2.rda | Bin 156 -> 151 bytes
data/rsmdat.rda | Bin 199 -> 196 bytes
inst/doc/Rplots.pdf | 4482 -------------------------------
inst/doc/UCML.jpg | Bin
inst/doc/Z.cls | 0
inst/doc/eRm.R | 101 -
inst/doc/eRm.Rnw | 3 +
inst/doc/eRm.pdf | Bin 555998 -> 469512 bytes
inst/doc/eRm_object_tree.pdf | Bin 0 -> 27586 bytes
inst/doc/eRmvig.R | 101 -
inst/doc/eRmvig.Rnw | 1006 -------
inst/doc/eRmvig.bib | 0
inst/doc/eRmvig.pdf | Bin 433168 -> 0 bytes
inst/doc/{index.html => index.html.old} | 0
inst/doc/jss.bst | 0
inst/doc/modelhierarchy.pdf | Bin
man/IC.Rd | 0
man/LLRA.Rd | 136 +
man/LLTM.Rd | 0
man/LPCM.Rd | 0
man/LRSM.Rd | 0
man/LRtest.Rd | 25 +-
man/MLoef.Rd | 21 +-
man/NPtest.Rd | 15 +-
man/PCM.Rd | 0
man/RM.Rd | 0
man/RSM.Rd | 0
man/Waldtest.Rd | 0
man/anova.llra.Rd | 65 +
man/build_W.Rd | 90 +
man/collapse_W.Rd | 76 +
man/eRm-package.Rd | 29 +-
man/gofIRT.Rd | 0
man/itemfit.ppar.Rd | 0
man/llra.datprep.Rd | 70 +
man/llraDat1.Rd | 61 +
man/llraDat2.Rd | 54 +
man/llradat3.Rd | 34 +
man/person.parameter.Rd | 0
man/plotDIF.Rd | 5 +-
man/plotGR.Rd | 57 +
man/plotICC.Rd | 0
man/plotPImap.Rd | 0
man/plotPWmap.Rd | 0
man/plotTR.Rd | 54 +
man/predict.ppar.Rd | 0
man/print.eRm.Rd | 0
man/raschdat.Rd | 0
man/sim.2pl.Rd | 0
man/sim.locdep.Rd | 0
man/sim.rasch.Rd | 0
man/sim.xdim.Rd | 0
man/stepwiseIt.Rd | 0
man/summary.llra.Rd | 69 +
man/thresholds.Rd | 0
src/components.c | 0
src/components.h | 0
src/geodist.c | 0
src/geodist.h | 0
180 files changed, 1654 insertions(+), 5925 deletions(-)
diff --git a/COPYING b/COPYING
old mode 100644
new mode 100755
diff --git a/COPYRIGHTS b/COPYRIGHTS
old mode 100644
new mode 100755
diff --git a/DESCRIPTION b/DESCRIPTION
old mode 100644
new mode 100755
index 5ae1f87..1b69213
--- a/DESCRIPTION
+++ b/DESCRIPTION
@@ -1,24 +1,28 @@
-Package: eRm
-Type: Package
-Title: Extended Rasch Modeling.
-Version: 0.13-4
-Date: 2011-03-23
-Author: Patrick Mair, Reinhold Hatzinger, Marco Maier
-Maintainer: Patrick Mair <patrick.mair at wu.ac.at>
-Description: eRm fits Rasch models (RM), linear logistic test models
- (LLTM), rating scale model (RSM), linear rating scale models
- (LRSM), partial credit models (PCM), and linear partial credit
- models (LPCM). Missing values are allowed in the data matrix.
- Additional features are the ML estimation of the person
- parameters, Andersen's LR-test, item-specific Wald test,
- Martin-Loef-Test, nonparametric Monte-Carlo Tests,
- itemfit and personfit statistics including infit and outfit
- measures, various ICC and related plots, automated stepwise
- item elimination, simulation module for various binary data
- matrices. An eRm platform is provided at R-forge (see URL).
-License: GPL
-URL: http://r-forge.r-project.org/projects/erm/
-Imports: graphics, stats, MASS, methods
-Depends: R (>= 2.12.0), splines, methods, RaschSampler
-LazyData: yes
-LazyLoad: yes
+Package: eRm
+Type: Package
+Title: Extended Rasch Modeling.
+Version: 0.14-0
+Date: 2011-06-05
+Author: Patrick Mair, Reinhold Hatzinger, Marco Maier
+Maintainer: Patrick Mair <patrick.mair at wu.ac.at>
+Description: eRm fits Rasch models (RM), linear logistic test models
+ (LLTM), rating scale model (RSM), linear rating scale models
+ (LRSM), partial credit models (PCM), and linear partial credit
+ models (LPCM). Missing values are allowed in the data matrix.
+ Additional features are the ML estimation of the person
+ parameters, Andersen's LR-test, item-specific Wald test,
+ Martin-Loef-Test, nonparametric Monte-Carlo Tests, itemfit and
+ personfit statistics including infit and outfit measures,
+ various ICC and related plots, automated stepwise item
+ elimination, simulation module for various binary data
+ matrices. An eRm platform is provided at R-forge (see URL).
+License: GPL
+URL: http://r-forge.r-project.org/projects/erm/
+Imports: graphics, stats, MASS, methods, Matrix
+Depends: R (>= 2.12.0), splines, methods, RaschSampler
+Suggests: lattice
+LazyData: yes
+LazyLoad: yes
+Packaged: 2011-06-05 11:22:28 UTC; hatz
+Repository: CRAN
+Date/Publication: 2011-06-06 07:03:22
diff --git a/MD5 b/MD5
new file mode 100644
index 0000000..cd4f507
--- /dev/null
+++ b/MD5
@@ -0,0 +1,183 @@
+94d55d512a9ba36caa9b7df079bae19f *COPYING
+d86eed66bba841658207fd609fdd3d4a *COPYRIGHTS
+c824379f1e7400c15cea5f79c12608b4 *DESCRIPTION
+2fa574614347ba33bf658a819b4fe98a *NAMESPACE
+c3686599805d8b0a1fcd21e60698b311 *NEWS
+d2d22f8b2176c97ca7df720f7a089794 *R/IC.default.R
+2bedf16f88332f159bfad9cdb36f4689 *R/IC.ppar.r
+2e4b547352dd7af179635bc46f9cdf87 *R/IC.r
+4bf43e52d582082bb387c70910851f74 *R/LLRA.R
+7826640f34c84fbb51764108b0961824 *R/LLTM.R
+bf407bc699aa2a8a6bb1e507cc32f049 *R/LPCM.R
+559e1ab0a83b6e78e7ae67ddb4a41b1e *R/LRSM.R
+e09054c9e41da89f00cf09d2ca0b67b2 *R/LRtest.R
+2d59903e184e948d35803dc9745d1044 *R/LRtest.Rm.R
+c4570ad02cb02e2263ff64507e5aa619 *R/MLoef.R
+85771f7f5668ecde577700138d893add *R/NPtest.R
+bbd93fc79991f98878c94fff0bcfaf30 *R/PCM.R
+136ebfb819fe8287b48d0d33c194815d *R/RM.R
+8f3a897135420e5fb99b1adebc4c8765 *R/ROCR_aux.R
+83e714c0d5bd37f8299292e3bf574a09 *R/RSM.R
+bc7be7af543fb355b3452bade6befcef *R/Rsquared.R
+f2527f30f48fcf4cfda14c9951186ead *R/Waldtest.R
+80e1701b16f045aaf746f28dc808c98e *R/Waldtest.Rm.R
+4de7e90e5eba8d4d90d7005bc6b9c38d *R/anova.llra.R
+b2e5064d8aa93edded5747f9f64f315c *R/build_W.R
+a8aa6e591ab2b76117211cdef4cdcf52 *R/checkdata.R
+40e9100130fde20aa73a04818d9909b3 *R/cldeviance.R
+013c3532078f0f62c326f5f7fe10e23e *R/cmlprep.R
+9f91686f144add619bbaaec474974fbb *R/coef.eRm.R
+0218542e5f566c443de1054b43d1dc6e *R/coef.ppar.R
+2a665e30eb32b2b8b9a7ee2cb846831b *R/collapse_W.R
+323a5c6fe9d49f18031aeeff22de0ee5 *R/confint.eRm.r
+a98e2b8f8c0dc9f184159e94c210611b *R/confint.ppar.r
+fd0612d832c6044df83689d42631a9af *R/confint.threshold.r
+b9da6dc9d83605723db2e81c69fd81b8 *R/cwdeviance.r
+9f6fff516f1257a642e18175e82ec53d *R/datcheck.LRtest.r
+4f7c2902266ffc2b21916d2791287ef9 *R/datcheck.R
+3ed2f9de5c5ec9a4d3dd66b40e0425d1 *R/datprep_LLTM.R
+b9868aa42173751e8d9fd8a1a16c49e1 *R/datprep_LPCM.R
+5bd2486b21ae4a9f196c7445aea4c958 *R/datprep_LRSM.R
+8eff0f65d596844d9810c3cfa58bc0a2 *R/datprep_PCM.R
+ca0af91c768a48fff0782a8b92d5ed9d *R/datprep_RM.R
+977fc96896e111fcf2692e274d9be346 *R/datprep_RSM.R
+39b6410834977409a957f4dc85fb8aaf *R/fitcml.R
+4878fc213dcd9d9a0b3b59353f06eebe *R/gofIRT.R
+57242fed1520c63de85bdc52e6a70c31 *R/gofIRT.ppar.R
+f9850156706dcad20dfedfc80c220b04 *R/hoslem.R
+71212f4c14003458fc3e30638f339629 *R/invalid.R
+e2acfcf0f69a54046e7500f37b11cce7 *R/itemfit.R
+f242e9d1dc269c42da573caeb5beb426 *R/itemfit.ppar.R
+d12c6019315fbc11dca8ce109e1883ab *R/labeling.internal.r
+c6cc30c6760a3e4fd1213ec55bd086e9 *R/likLR.R
+9c56aa9ca79069fd2fa2a02f14d8e7b4 *R/llra.datprep.R
+b120a9958f01935b8829c7473345e9de *R/llra.internals.R
+5bea9ef9bfb310d9d57960b335cc6e2f *R/logLik.eRm.r
+6f889b8ad497988f236902afa009cba3 *R/logLik.ppar.r
+a4b58be8d00e8a61a442e6e7f5709138 *R/model.matrix.eRm.R
+08c4ee8081fb13b7da90b8a6faa28ea0 *R/performance.R
+ad14b6669d085f886f032ff60fd2643b *R/performance_measures.R
+166e801703a416aa2a2cdbde298651db *R/performance_plots.R
+adabedf84b1cecfce267c7be91d7cbbb *R/person.parameter.R
+e91f0d25dcaa66df375fa34312e0f36c *R/person.parameter.eRm.R
+1580451632297ddae0e4ae6c262a9bb0 *R/personfit.R
+68cc2204b1c470c10a57948e6629d542 *R/personfit.ppar.R
+5211c838890a01fe791e26e5f2ba10c8 *R/pifit.internal.r
+6dc484eac9810ce58aa2a06fab305a56 *R/plist.internal.R
+e050e7af6f9bb0a52bf55026d3101d27 *R/plot.ppar.r
+fa5b8a3513fcdf05e0533c7549d7adb7 *R/plotCI.R
+5b20ac29faf764d0e70892b578012162 *R/plotDIF.R
+a8afc0168d588f3380174f8e7b6d37cd *R/plotGOF.LR.R
+c535a897ffcc871b893cbbee237baf70 *R/plotGOF.R
+c0c259828a7490f7cf87888394e31214 *R/plotGR.R
+e5df0fcbc0cf84479dd1af053994135c *R/plotICC.R
+0add981ba89f86649c26ca8c5bd818b6 *R/plotICC.Rm.R
+e9b4f3c4e961531fbd204c4e52e056a4 *R/plotPImap.R
+53328e5a237f91c7f03e1447430a3580 *R/plotPWmap.R
+9f9956d4213bf12105e1985ff7c2fc28 *R/plotTR.R
+817427f50be483fdf3dc5be097121c7f *R/plotjointICC.R
+30e0c7cadcb3d360f9373b6ffdc58912 *R/plotjointICC.dRm.R
+503d39882ff1519f9a9c822525df881b *R/pmat.R
+978c189f51d4c525f6a1d53377c72fab *R/pmat.default.R
+3b8b125515c56d967e0184d9bae4fe87 *R/pmat.ppar.R
+75667db3c392cce69583eb02f93739bb *R/predict.ppar.R
+c6520d9937a5432bba0921b1c1ddc154 *R/prediction.R
+2ef33bcb8ff18bada70cc9cd28167b6b *R/print.ICr.r
+e419bb90c13b3e34d60f809d2aab7a8c *R/print.LR.R
+baa854c72760319d47628d9cdeab5847 *R/print.MLoef.r
+063e72adfa562645ff3510ff2f93fc8a *R/print.eRm.R
+c5312a2efff02316933e78574a2d7072 *R/print.gof.R
+1a8e26e12c84b76aac1cccc7ecafd6c0 *R/print.ifit.R
+eceef6e388090c985b69bb7495fd898b *R/print.llra.R
+37817e03ebd0e3395fb011928892e97b *R/print.logLik.eRm.R
+e7e35e2b4d5cfd70f341b852793c0e4b *R/print.logLik.ppar.r
+ae0c91d5cdfde5d20945fc0e804708f7 *R/print.pfit.R
+22c7359fd707bd836a258c6e02f484b3 *R/print.ppar.R
+cb2c9116b9ede352bb567fa9eb8ac393 *R/print.resid.R
+45ce31571c765ec1a1bb87a16ff82184 *R/print.step.r
+22252f71f8a50bf9726528501c70b443 *R/print.summary.llra.R
+5aa80f82e5e72c24602c8525b6389d71 *R/print.threshold.r
+2ba6872ce486a3b88889f50ab34046f1 *R/print.wald.R
+681692c0b23c20c07fd6e27eb6823d03 *R/residuals.ppar.R
+7a34777d662d5c0f5c9c1e473bf52746 *R/rostdeviance.r
+b3f9c25354a6b9bddf925bcab2648687 *R/sim.2pl.R
+4b54d34fbb18f74fac3b93da2c11a84f *R/sim.locdep.R
+f882b1817e7bbcef9b27c276f257a626 *R/sim.rasch.R
+60c5d7095026f9b3f7ee2dc6321fbc25 *R/sim.xdim.R
+50fba0ca19951546155a7d644228d24f *R/stepwiseIt.R
+ec78daaa3f45ed12a62e483f3ee1ce63 *R/stepwiseIt.eRm.R
+913ccfc27b8531d6efe242b877b45d30 *R/summary.LR.r
+65d0e9b84e5f96d02e9de47a6e8a5a90 *R/summary.MLoef.r
+dca2b5cf6a23b084d1fcee50a51050b6 *R/summary.eRm.R
+a201d1040cd729f62e09a45ff0402957 *R/summary.gof.R
+91593cc56050024103f88648bd3fd9d6 *R/summary.llra.R
+bde67b4f83ca342816e87e9c39fde213 *R/summary.ppar.R
+d28668718027ba4f7d4210dcac0c6b76 *R/summary.threshold.r
+b80b7dbe79cca5167e85e53674887e40 *R/thresholds.eRm.r
+3c6f6ec631aeacf22e73fadf8074fc12 *R/thresholds.r
+54259c5861ddc4a41474b0f47dc7ff13 *R/vcov.eRm.R
+e7182e7050856376cc68a090c3fc962d *R/zzz.R
+17e5211154d381fb70abc1eddbccb9d7 *data/llraDat1.rda
+04e416237eb526a153a793cec2988809 *data/llraDat2.rda
+9346da0bb55f95db5ed71f73d050f2bd *data/llradat3.rda
+ed5e5d66de298a34f143f361ee0722fa *data/lltmdat1.rda
+b66d20feef24d2792b7e496bf30dbb7e *data/lltmdat2.rda
+d064d228919bf4bec3f96f3bdc73309a *data/lpcmdat.rda
+1ae66d8c67c396f15799b2b8defd9447 *data/lrsmdat.rda
+edd14b4be4ee7329f78e2ec1555a4ad4 *data/pcmdat.rda
+3694b4d3076bbbb26f6734c06e923f90 *data/pcmdat2.rda
+257737a046f48862ca6f26fc8f5bc94f *data/raschdat1.rda
+18d2da8d7902349b513004d1342fbd3d *data/raschdat2.rda
+c42bff96b6fd7f98d0b86bd6b058b367 *data/rsmdat.rda
+c58787b71a1f1d3c295a4bfa20737306 *inst/doc/UCML.jpg
+859438dab150c3e7762183e79cb4e6ec *inst/doc/Z.cls
+6c088c475da8ad1998ff6abf0550a717 *inst/doc/eRm.Rnw
+53f32799d25b50133b6ff217357184ae *inst/doc/eRm.pdf
+61d680540f93d252561fa615bd95511b *inst/doc/eRm_object_tree.pdf
+4252672da961d1370a22bc901d68cc5d *inst/doc/eRmvig.bib
+405e1ee0b6273a4d992299ddc0f9233c *inst/doc/index.html.old
+92e903f33d4067a7fbc89fa4e7571c92 *inst/doc/jss.bst
+e97dfac8265ca8a3cbae7ff1d64ac832 *inst/doc/modelhierarchy.pdf
+3ccae9025b31a4c6c2a4a0c38fadcc60 *man/IC.Rd
+346aa4dad160d49be05a16a4f6fae983 *man/LLRA.Rd
+3ff5fc2b6804ccec50788a81ab024ad3 *man/LLTM.Rd
+b2674dd262ed08acf9973e028b6c9062 *man/LPCM.Rd
+2fe4a48d3db49e43f8ca765b3724a23b *man/LRSM.Rd
+21a5eee857676dadca51fce6f58438bb *man/LRtest.Rd
+145b3975b694fcb6040bb7ece98b4962 *man/MLoef.Rd
+07664ccf23e9418d1a3e2ae70572aa82 *man/NPtest.Rd
+49e40371caf75ce2b50dde4fce78fa83 *man/PCM.Rd
+ba1bc663be054dca6f66956e83296d9f *man/RM.Rd
+81a06eda5b851e2551d0c069132bd601 *man/RSM.Rd
+1fa68e025a4ece034a6e9719c58f26e7 *man/Waldtest.Rd
+7a672a0bdbee1d6dfcddd76a77e7f384 *man/anova.llra.Rd
+61d45eca1d3602e0910c562444a5a91a *man/build_W.Rd
+5646ca17db14129c73ecc8e2c059f0d6 *man/collapse_W.Rd
+3df6fa3e7648ed587015d46e61ff43f3 *man/eRm-package.Rd
+e61327803e5e6b7d2d3ccc944e37aabb *man/gofIRT.Rd
+76cea65dee359343eacaf632bda3ea52 *man/itemfit.ppar.Rd
+07fa1d3822aa64ac126dd6ea09fbd896 *man/llra.datprep.Rd
+3b7d3041daf6dff31a863240a7d66c43 *man/llraDat1.Rd
+6c2f252c609fbf7bf5e909c7aaae01f3 *man/llraDat2.Rd
+4e6086c12e5f1af87effae54623069b8 *man/llradat3.Rd
+5858e38d4bd68f93c3eb4961423f2aae *man/person.parameter.Rd
+cbae5267dff3868ef0cd2c382a9968da *man/plotDIF.Rd
+cc89a7415fd66ead9a764e66e190ebb8 *man/plotGR.Rd
+ebc0198383f388c7f118bf004e1ab47a *man/plotICC.Rd
+fb7839cc73bd28943352b8af8601cb87 *man/plotPImap.Rd
+db3395d4738132591b935af5dc7809ac *man/plotPWmap.Rd
+bb2e316fb66fbc62d5d9d86db9ced96a *man/plotTR.Rd
+5c7fab3317a8fcc01c2ff0a6fe1824a9 *man/predict.ppar.Rd
+9e66621272c2cf652b1396428edf94f8 *man/print.eRm.Rd
+d96cb1ddd85fd3b7ecdfe095550dc027 *man/raschdat.Rd
+8cd62fbd653f0036bdc4ff0a97c6c77d *man/sim.2pl.Rd
+91b5aef01c0f9142c7970c6b75727034 *man/sim.locdep.Rd
+b937031cca9d299b55d3932f299983af *man/sim.rasch.Rd
+242e82abb41d92f5926ec9ff2b037e1b *man/sim.xdim.Rd
+77f43c8fcfd4ff3af7f4351419fde0ff *man/stepwiseIt.Rd
+a62ad6ab0c17c4af3ea1de449c0cc388 *man/summary.llra.Rd
+eb6aee99b123cc957f76c938c624f264 *man/thresholds.Rd
+d0d66f1a61eae0e016a6e9401d0e5917 *src/components.c
+ce4282b827566f0ef0572ae06eb7bf04 *src/components.h
+b318d7bfff0ba1eccadc86d34e1f0c5f *src/geodist.c
+c44d6148a344eb9e3deb558f5d453d8c *src/geodist.h
diff --git a/NAMESPACE b/NAMESPACE
old mode 100644
new mode 100755
index cccc505..d205143
--- a/NAMESPACE
+++ b/NAMESPACE
@@ -1,5 +1,6 @@
useDynLib(eRm)
import("stats", "graphics")
+importFrom(Matrix, bdiag)
export(RM)
export(LLTM)
export(RSM)
@@ -28,6 +29,12 @@ export(sim.locdep)
export(stepwiseIt)
export(gofIRT)
export(NPtest)
+export(llra.datprep)
+export(build_W)
+export(collapse_W)
+export(LLRA)
+export(plotGR)
+export(plotTR)
S3method(print, eRm)
S3method(summary, eRm)
@@ -39,6 +46,7 @@ S3method(coef, eRm)
S3method(coef, ppar)
S3method(vcov, eRm)
S3method(print, LR)
+S3method(print, MLobj)
S3method(print, MLoef)
S3method(print, ifit)
S3method(print, wald)
@@ -80,3 +88,7 @@ S3method(print, T7obj)
S3method(print, T7aobj)
S3method(print, T10obj)
S3method(print, T11obj)
+S3method(anova, llra)
+S3method(print, llra)
+S3method(print,summary.llra)
+S3method(summary, llra)
diff --git a/NEWS b/NEWS
old mode 100644
new mode 100755
index c1f3c5c..8e2a01b
--- a/NEWS
+++ b/NEWS
@@ -1,12 +1,23 @@
-Changes in Version 0.13-4
+Changes in Version 0.14-0
+
+ o new (wrapper) function LLRA for fitting linear logistic
+ models with relaxed assumptions including utilities for
+ preparing data (llra.datprep), setting up (build_W) and
+ modifying (collapse_W) design matrices, comparing llra
+ models (anova) and plotting results (plotTR and plotGR)
+
+ o 'exact' version of the Martin-Loef test for binary items and
+ arbitrary splits added as method to NPtest
+
+ o in plotGOF confidence ellipses can now be drawn for
+ subsets of items, optionally using different colours
o new function plotDIF (by Kathrin Gruber): plots confidence
intervals for item parameters estimated separately in
subgroups, uses LR objects as input
-
- o adapted the MLoef function to work with polytomous data
-Changes in Version 0.13-3
+ o adapted the MLoef function to work with polytomous data
+ and more than two item groups
o error checks in NPtest: (i) 0/full resposes for items
meaningless for NPtest, (ii) group in method="T4" must
@@ -16,8 +27,6 @@ Changes in Version 0.13-3
o warning regarding group assignment when using median
split removed from LRtest and Waldtest
-Changes in Version 0.13-2
-
o some modifications in plotPWmap: horizontal plotting,
different default plotting symbols, option to change
size of plotting symbols
@@ -29,9 +38,6 @@ Changes in Version 0.13-2
o Latin1 encoding removed
-
-Changes in Version 0.13-1
-
o bug in plotICC (always same title) fixed
Changes in Version 0.13-0
diff --git a/R/IC.default.R b/R/IC.default.R
old mode 100644
new mode 100755
diff --git a/R/IC.ppar.r b/R/IC.ppar.r
old mode 100644
new mode 100755
diff --git a/R/IC.r b/R/IC.r
old mode 100644
new mode 100755
diff --git a/R/LLRA.R b/R/LLRA.R
new file mode 100755
index 0000000..557dfb8
--- /dev/null
+++ b/R/LLRA.R
@@ -0,0 +1,16 @@
+LLRA <- function(X, W, mpoints, groups, baseline=NULL, itmgrps=NULL,...)
+{
+ if(missing(mpoints)) stop("Please specify the number of time points. If there are none, you might want to try PCM() or LPCM().")
+ Xprep <- llra.datprep(X,mpoints,groups,baseline)
+ itmgrps <- rep(1:Xprep$nitems)
+ groupvec <- Xprep$assign.vec
+ pplgrps <- length(Xprep$grp_n)
+ if(missing(W)) W <- build_W(Xprep$X,length(unique(itmgrps)),mpoints,Xprep$grp_n,groupvec,itmgrps)
+ fit <- LPCM(Xprep$X,W,mpoints=mpoints,groupvec=groupvec,sum0=FALSE)
+ refg <- unique(names(which(groupvec==max(groupvec))))
+ out <- c(fit,"itms"=Xprep$nitems,"refGroup"=refg)
+ out$call <- match.call()
+ class(out) <- c("llra","Rm","eRm")
+ cat("Reference group: ",refg,"\n\n")
+ return(out)
+}
diff --git a/R/LLTM.R b/R/LLTM.R
old mode 100644
new mode 100755
diff --git a/R/LPCM.R b/R/LPCM.R
old mode 100644
new mode 100755
diff --git a/R/LRSM.R b/R/LRSM.R
old mode 100644
new mode 100755
diff --git a/R/LRtest.R b/R/LRtest.R
old mode 100644
new mode 100755
diff --git a/R/LRtest.Rm.R b/R/LRtest.Rm.R
old mode 100644
new mode 100755
diff --git a/R/MLoef.R b/R/MLoef.R
old mode 100644
new mode 100755
diff --git a/R/NPtest.R b/R/NPtest.R
old mode 100644
new mode 100755
index 6a7e4a2..94a88d6
--- a/R/NPtest.R
+++ b/R/NPtest.R
@@ -13,7 +13,6 @@ NPtest<-function(obj, n=NULL, method="T1", ...){
if (is.null(n)) n <- 500
obj<-rsampler(obj,rsctrl(burn_in=256, n_eff=n, step=32))
}
-
switch(method,
"T1"=T1(obj),
"T2"=T2(obj, ...),
@@ -21,10 +20,35 @@ NPtest<-function(obj, n=NULL, method="T1", ...){
"T7"=T7(obj, ...),
"T7a"=T7a(obj, ...),
"T10"=T10(obj, ...),
- "T11"=T11(obj)
+ "T11"=T11(obj),
+ "MLoef"=MLoef.x(obj, ...) ###############################################
)
}
+MLoef.x<-function(rsobj, splitcr=NULL){
+ # user function
+ MLexact<-function(X,splitcr){
+ rmod<-RM(X)
+ LR<-MLoef(rmod,splitcr)$LR
+ LR
+ }
+ #if(!exists("splitcr")) splitcr="median"
+ if(is.null(splitcr)) splitcr="median"
+ res <- rstats(rsextrobj(rsobj, 2), MLexact, splitcr)
+
+ rmod<-RM(rsextrmat(rsobj,1)) # MLoef for original data
+ MLres<-MLoef(rmod,splitcr)
+ class(MLres)<-c(class(MLres),"MLx") # for printing without blank line
+ res1<-MLres$LR
+
+ n_eff<-rsobj$n_eff # number of simulated matrices
+ res<-unlist(res)
+ prop<-sum((res[1:n_eff]>=res1)/n_eff)
+
+ result<-list(MLres=MLres, n_eff=n_eff, prop=prop, MLoefvec=res) # MLobj
+ class(result)<-"MLobj"
+ result
+}
T1<-function(rsobj){
T1stat<-function(x){ # calculates statistic T1
unlist(lapply(1:(k-1),function(i) lapply((i+1):k, function(j) sum(x[,i]==x[,j]))))
@@ -275,13 +299,18 @@ T11<-function(rsobj){
result
}
+print.MLobj<-function(x,...){
+ print(x$MLres)
+ cat("'exact' p-value =", x$prop, " (based on", x$n_eff, "sampled matrices)\n\n")
+}
+
print.T1obj<-function(x,alpha=0.05,...){
txt1<-"\nNonparametric RM model test: T1 (local dependence - increased inter-item correlations)\n"
writeLines(strwrap(txt1, exdent=5))
- cat(" (counting cases with equal responses on both items)\n\n")
+ cat(" (counting cases with equal responses on both items)\n")
cat("Number of sampled matrices:", x$n_eff,"\n")
- cat("\nNumber of Item-Pairs tested:", length(x$prop),"\n")
- cat("\nItem-Pairs with one-sided p <", alpha,"\n\n")
+ cat("Number of Item-Pairs tested:", length(x$prop),"\n")
+ cat("Item-Pairs with one-sided p <", alpha,"\n")
T1mat<-x$T1mat
idx<-which(T1mat<alpha,arr.ind=TRUE)
val<-T1mat[which(T1mat<alpha)]
@@ -303,11 +332,11 @@ print.T2obj<-function(x,...){
)
txt<-"\nNonparametric RM model test: T2 (local dependence - model deviating subscales)\n"
writeLines(strwrap(txt, exdent=5))
- cat(" (dispersion of subscale person rawscores)\n\n")
+ cat(" (dispersion of subscale person rawscores)\n")
cat("Number of sampled matrices:", x$n_eff,"\n")
cat("Items in subscale:", idx,"\n")
cat("Statistic:", statnam,"\n")
- cat("one-sided p-value:",prop,"\n")
+ cat("one-sided p-value:",prop,"\n\n")
# cat(" (proportion of sampled",statnam," GE observed)\n\n")
}
print.T4obj<-function(x,...){
@@ -317,7 +346,7 @@ print.T4obj<-function(x,...){
gr.n<-x$gr.n
alternative<-x$alternative
cat("\nNonparametric RM model test: T4 (Group anomalies - DIF)\n")
- cat(" (counting", alternative, "raw scores on item(s) for specified group)\n\n")
+ cat(" (counting", alternative, "raw scores on item(s) for specified group)\n")
cat("Number of sampled matrices:", x$n_eff,"\n")
cat("Items in Subscale:", idx,"\n")
cat("Group:",gr.nam," n =",gr.n,"\n")
@@ -328,16 +357,16 @@ print.T4obj<-function(x,...){
print.T7obj<-function(x,...){
prop<-x$prop
cat("\nNonparametric RM model test: T7 (different discrimination - 2PL)\n")
- cat(" (counting cases with response 1 on more difficult and 0 on easier item)\n\n")
+ cat(" (counting cases with response 1 on more difficult and 0 on easier item)\n")
cat("Number of sampled matrices:", x$n_eff,"\n")
- cat("\nItem Scores:\n")
+ cat("Item Scores:\n")
print(x$itscor)
cat("one-sided p-value:",prop,"\n\n")
}
print.T7aobj<-function(x,...){
prop<-x$prop
cat("\nNonparametric RM model test: T7a (different discrimination - 2PL)\n")
- cat(" (counting cases with response 1 on more difficult and 0 on easier item)\n\n")
+ cat(" (counting cases with response 1 on more difficult and 0 on easier item)\n")
cat("Number of sampled matrices:", x$n_eff,"\n")
cat("Item Scores:\n")
print(x$itscor)
@@ -349,7 +378,7 @@ print.T10obj<-function(x,...){
prop<-x$prop
hi.n<-x$hi.n
low.n<-x$low.n
- cat("\nNonparametric RM model test: T10 (global test - subgroup-invariance)\n\n")
+ cat("\nNonparametric RM model test: T10 (global test - subgroup-invariance)\n")
cat("Number of sampled matrices:", x$n_eff,"\n")
cat("Split:",spl.nam,"\n")
cat("Group 1: n = ",hi.n," Group 2: n =",low.n,"\n")
@@ -359,7 +388,7 @@ print.T10obj<-function(x,...){
print.T11obj<-function(x,...){
prop<-x$prop
cat("\nNonparametric RM model test: T11 (global test - local dependence)\n")
- cat(" (sum of deviations between observed and expected inter-item correlations)\n\n")
+ cat(" (sum of deviations between observed and expected inter-item correlations)\n")
cat("Number of sampled matrices:", x$n_eff,"\n")
cat("one-sided p-value:",prop,"\n\n")
# cat(" (proportion of sampled sums GE observed)\n\n")
diff --git a/R/PCM.R b/R/PCM.R
old mode 100644
new mode 100755
diff --git a/R/RM.R b/R/RM.R
old mode 100644
new mode 100755
diff --git a/R/ROCR_aux.R b/R/ROCR_aux.R
old mode 100644
new mode 100755
diff --git a/R/RSM.R b/R/RSM.R
old mode 100644
new mode 100755
diff --git a/R/Waldtest.R b/R/Waldtest.R
old mode 100644
new mode 100755
diff --git a/R/Waldtest.Rm.R b/R/Waldtest.Rm.R
old mode 100644
new mode 100755
diff --git a/R/anova.llra.R b/R/anova.llra.R
new file mode 100755
index 0000000..b815842
--- /dev/null
+++ b/R/anova.llra.R
@@ -0,0 +1,31 @@
+anova.llra <- function(object, ...) UseMethod("anova.llra")
+
+anova.llra.default <- function(object,...)
+ {
+ objets <- list(object, ...)
+ isllra <- unlist(lapply(objets, function(x) "llra" %in% class(x)))
+
+ ## checks
+ if (!all(isllra)) {
+ objets <- objets[isllra]
+ warning("non-LLRA-model(s) removed")
+ }
+ dimdata <- dim(objets[[1L]]$X)
+ samedata <- unlist(lapply(objets, function(x) all(dim(x$X)==dimdata)))
+ if (!all(samedata))
+ stop("models were not all fitted to the same size of dataset")
+
+ nmodels <- length(objets)
+ logliks <- as.numeric(lapply(objets, function(x) x$loglik))
+ npars <- as.numeric(lapply(objets, function(x) x$npar))
+ nparsS <- npars[order(npars)]
+ logliksS <- logliks[order(npars)]
+ lrstat <-c(NA,2*abs(diff(logliksS)))
+ dfs <- c(NA,abs(diff(nparsS)))
+ ps <- 1-pchisq(lrstat,dfs)
+ tbl <- data.frame(nparsS, logliksS, dfs, lrstat, ps)
+ dimnames(tbl) <- list(1L:nmodels, c("Npar", "logLik", "df", "-2LR","Asymp.p-Value"))
+ title <- "Analysis of Deviance Table\n"
+ structure(tbl, heading = title, class = c("anova",
+ "data.frame"))
+}
diff --git a/R/build_W.R b/R/build_W.R
new file mode 100755
index 0000000..79fa417
--- /dev/null
+++ b/R/build_W.R
@@ -0,0 +1,21 @@
+build_W <- function(X,nitems,mpoints,grp_n,groupvec,itmgrps)
+ {
+ if(missing(grp_n)) grp_n<- table(groupvec)
+ if(!is.numeric(grp_n)) stop("Please specify the number of subjects per group.")
+ if(missing(nitems))stop("Please specify the number of items.")
+ if(any(grp_n==0)) stop("There are groups with zero sample size.")
+ if(missing(mpoints)) stop("Please specify the number of time points. If there are none, you might want to use PCM() or LPCM().")
+ pplgrps <- length(grp_n)
+ #builds the LLRA design matrix from scratch
+ categos <- get_item_cats(X,nitems,grp_n)
+ #trend effects design
+ tr.des <- build_trdes(nitems,mpoints,pplgrps,categos)
+ #tretment effects design
+ gr.des <- build_effdes(nitems,mpoints,pplgrps,categos,groupvec)
+ #category design
+ if(length(unique(unlist(categos)))==1&&sum(unique(unlist(categos)))==1) return(cbind(gr.des,tr.des))
+ ct.des <- build_catdes(nitems,mpoints,pplgrps,categos)
+ #all together now!
+ des <-cbind(gr.des,tr.des,ct.des)
+ des
+ }
diff --git a/R/checkdata.R b/R/checkdata.R
old mode 100644
new mode 100755
diff --git a/R/cmlprep.R b/R/cmlprep.R
old mode 100644
new mode 100755
diff --git a/R/coef.eRm.R b/R/coef.eRm.R
old mode 100644
new mode 100755
diff --git a/R/coef.ppar.R b/R/coef.ppar.R
old mode 100644
new mode 100755
diff --git a/R/collapse_W.R b/R/collapse_W.R
new file mode 100755
index 0000000..136ef7b
--- /dev/null
+++ b/R/collapse_W.R
@@ -0,0 +1,10 @@
+collapse_W <- function(W,listItems,newNames)
+ {
+ if(missing(newNames)) newNames <- paste("collapsedEffect",1:length(listItems),sep="")
+ Wtmp1 <- W[,-unlist(listItems)]
+ collapsed <- lapply(listItems, function(x) rowSums(W[,x]))
+ Wtmp2 <- matrix(unlist(collapsed),ncol=length(collapsed))
+ Wout <- cbind(Wtmp1,Wtmp2)
+ colnames(Wout) <- c(colnames(Wtmp1),newNames)
+ Wout
+ }
diff --git a/R/confint.eRm.r b/R/confint.eRm.r
old mode 100644
new mode 100755
diff --git a/R/confint.ppar.r b/R/confint.ppar.r
old mode 100644
new mode 100755
diff --git a/R/confint.threshold.r b/R/confint.threshold.r
old mode 100644
new mode 100755
diff --git a/R/cwdeviance.r b/R/cwdeviance.r
old mode 100644
new mode 100755
diff --git a/R/datcheck.LRtest.r b/R/datcheck.LRtest.r
old mode 100644
new mode 100755
diff --git a/R/datcheck.R b/R/datcheck.R
old mode 100644
new mode 100755
diff --git a/R/datprep_LLTM.R b/R/datprep_LLTM.R
old mode 100644
new mode 100755
diff --git a/R/datprep_LPCM.R b/R/datprep_LPCM.R
old mode 100644
new mode 100755
diff --git a/R/datprep_LRSM.R b/R/datprep_LRSM.R
old mode 100644
new mode 100755
diff --git a/R/datprep_PCM.R b/R/datprep_PCM.R
old mode 100644
new mode 100755
diff --git a/R/datprep_RM.R b/R/datprep_RM.R
old mode 100644
new mode 100755
diff --git a/R/datprep_RSM.R b/R/datprep_RSM.R
old mode 100644
new mode 100755
diff --git a/R/fitcml.R b/R/fitcml.R
old mode 100644
new mode 100755
diff --git a/R/invalid.R b/R/invalid.R
old mode 100644
new mode 100755
diff --git a/R/itemfit.R b/R/itemfit.R
old mode 100644
new mode 100755
diff --git a/R/itemfit.ppar.R b/R/itemfit.ppar.R
old mode 100644
new mode 100755
diff --git a/R/labeling.internal.r b/R/labeling.internal.r
old mode 100644
new mode 100755
diff --git a/R/likLR.R b/R/likLR.R
old mode 100644
new mode 100755
diff --git a/R/llra.datprep.R b/R/llra.datprep.R
new file mode 100755
index 0000000..67cdd5a
--- /dev/null
+++ b/R/llra.datprep.R
@@ -0,0 +1,43 @@
+llra.datprep<-function(X, mpoints, groups, baseline=NULL){
+
+ Xwide <- X
+ if (ncol(Xwide) %% mpoints > 0)
+ stop("Number of items must be the same for each timepoint.")
+ nitems <- dim(Xwide)[2]/mpoints
+
+ if(missing(groups)) groups <- rep("CG",dim(Xwide)[1])
+
+ covs.prep<-function(groups,baseline){
+ groups<-as.matrix(groups)
+ grstr<-apply(groups,1,paste,collapse=":")
+ grstr <- factor(grstr)
+ if(!is.null(baseline))
+ {
+ basel <- paste(baseline,collapse=":")
+ grstr <- relevel(grstr,basel)
+ }
+ cov.groupvec<-as.numeric(grstr)
+ names(cov.groupvec)<-grstr
+ cov.groupvec
+ }
+
+ # sort data according to cov.groupvec
+ cov.groupvec<-covs.prep(groups,baseline)
+ Xwide<-Xwide[order(cov.groupvec,decreasing=TRUE),]
+ cov.groupvec<-sort(cov.groupvec)
+
+ # number of people per group
+ grp_n<-table(cov.groupvec)
+ names(grp_n)<-unique(names(cov.groupvec))
+
+ # convert to long format
+ Xlong <- matrix(unlist(Xwide), ncol = mpoints)
+
+ # assignment vector item x treatment
+ assign.vec <- as.vector(sapply(1:nitems, function(i)
+ cov.groupvec + (i-1)*max(cov.groupvec)))
+ assign.vec <- rev(assign.vec)
+ assign.vec <- abs(assign.vec-max(assign.vec))+1
+ names(assign.vec)<-rev(rep(names(cov.groupvec),nitems))
+ list(X=Xlong, assign.vec=assign.vec, grp_n=grp_n, nitems=nitems)
+}
diff --git a/R/llra.internals.R b/R/llra.internals.R
new file mode 100755
index 0000000..a9195ca
--- /dev/null
+++ b/R/llra.internals.R
@@ -0,0 +1,107 @@
+#internal functions
+get_item_cats <- function(X,nitems,grp_n)
+ {
+ # returns list of vectors with length max(categories) for each item;
+ # 1:number categories are the first few entries and the rest is filed with zeros
+ # This later corresponds to the necessary setup in LPCM where the superfluous categories must be set to 0
+ its <- rep(1:nitems,each=sum(grp_n))
+ cats <- lapply(split(X,its),max) #splits the data matrix according to items and finds the maximum category
+ max.cat <- max(X) #overall maximum category
+ vec.cat <- lapply(cats,function(x) c(1:x,rep(0,max.cat-x)))
+ vec.cat #the ominous list of form c(1:categories,0,0,0)
+ }
+
+build_effdes <- function(nitems,mpoints,pplgrps,categos,groupvec)
+ {
+ #builds treatment design structure for W
+ #
+ #mpoints>nitems>treat>catego
+ #build group design
+ tmp1 <- diag(pplgrps)
+ tmp1[pplgrps,pplgrps] <- 0
+ eff.tmp1 <- lapply(categos,function(x)(tmp1%x%x)) #list with categories per item, replicated per group
+ eff.tmp2 <- as.matrix(bdiag(eff.tmp1)) #blockdiagonal with blocks equal to the categories
+ eff.tmp3 <- diag(mpoints-1)%x%eff.tmp2 #blow up to mpoints
+ nuller <- matrix(0,nrow=dim(eff.tmp2)[1],ncol=dim(eff.tmp3)[2]) #baseline (tp=1)
+ gr.bu <- rbind(nuller,eff.tmp3) #combine baseline and effects
+ #labelling of effects
+ names1 <- unique(names(groupvec))
+ #names1 <- paste("G",pplgrps:1,sep="")
+ names2 <- paste(names1,"I",sep=".")
+ names3 <- paste(names2,rep(1:nitems,each=pplgrps),sep="")
+ names4 <- paste(names3,"t",sep=".")
+ names5 <- paste(names4,rep(2:mpoints,each=pplgrps*nitems),sep="")
+ colnames(gr.bu) <- names5
+ #columns with zeros (baseline group) are removed now
+ rem.0 <- NA
+ for(i in 1:dim(gr.bu)[2]) {rem.0[i] <- all(gr.bu[,i]==0)}
+ gr.bu.red <- gr.bu[,which(rem.0==0)]
+ return(gr.bu.red)
+ }
+
+
+build_trdes <- function(nitems,mpoints,pplgrps,categos)
+ {
+ #builds trend design structure for W
+ #
+ #mpoints>nitems>treat>catego
+ tr.tmp1 <- lapply(categos,function(x) rep(x,pplgrps)) #replicate number of categories per item times the groups
+ tr.tmp2 <- as.matrix(bdiag(tr.tmp1)) #build the blockdiaginal for all items
+ tr.tmp3 <- diag(mpoints-1)%x%tr.tmp2 #blow it up to the time points necessary
+ nuller <- matrix(0,nrow=dim(tr.tmp2)[1],ncol=dim(tr.tmp3)[2]) #baseline
+ tr.bu <- rbind(nuller,tr.tmp3) #combine mpoints and baseline
+ #structure: for each category multiply it with a vector of group indicators
+ #hence the grouping is:
+ #tau1 t2-t1, tau2 t2-t1, ..., tauk t2-t1, tau1 t3-t1, tau2 t3-t1, .. tauk t3-t1
+ #cat("Design matrix columns are:","\n","tau_1^(t2-t1), tau_2^(t2-t1), ..., tau_k^(t2-t1), tau_1^(t3-t1), tau_2(t3-t1), ..., tau_k^(t3-t1), etc.","\n")
+ #labeling
+ names1 <- paste("trend.I",1:nitems,sep="")
+ names2 <- paste(names1,"t",sep=".")
+ names3 <- paste(names2,rep(2:mpoints,each=nitems),sep="")
+ colnames(tr.bu) <- names3
+ return(tr.bu)
+ }
+
+build_catdes <- function(nitems,mpoints,pplgrps,categos)
+ {
+ #builds category design matrix
+ #FIX ME: is a bit ugly, we might get the loops out somehow
+ #
+ #check if there are just binary items
+ if(max(unlist(categos))<2) stop("items are (at most) binary and need no design")
+ #currently equates cat.0 and cat.1
+ warning("Currently c0 and c1 are equated for each item","\n")
+ max.all <- max(unlist(categos)) #maximum category number
+ ls.ct.des <- list() #list of designs for each item
+ #here we walk through each item and build up the category design
+ for(i in 1:nitems)
+ {
+ max.it <- sum(categos[[i]]!=0) #maximum category number of item i
+ ct.des <- rbind(rep(0,dim(diag(max.all-1))[2]),diag(max.all-1)) #the design for the maximum number of categories in X
+ rems <- max.all-max.it #the number of superfluous columns
+ #here the superfluous columns are removed as the step from W to W*
+ #the necessary rows with zeros however are maintained:
+ #for a dichotomous item the structure is slightly different than for any other, since it returns an empty matrix of appropriate dimensions
+ #for a polytomous item the superfluous columns are removed from the back
+ ifelse(rems==max.all-1, ct.des<- as.matrix(ct.des[,-(1:max.all-1)]), ct.des<- as.matrix(ct.des[,1:((max.all-1)-rems)]))
+ ct.des.gr <- rep(1,pplgrps)%x%ct.des #blow it up to the number of groups
+ ls.ct.des[[i]] <- ct.des.gr #list with all category designs for each item
+ }
+ ct.tmp2 <- as.matrix(bdiag(ls.ct.des)) #blockdiagonal matrix for a single mpoints
+ ct.bu <- rep(1,mpoints)%x%ct.tmp2 #blow up to number of times points
+ #try to first build first item, then second and so on, then blow up
+ #labeling: pretty unelegant too
+ names <- NA
+ for(i in 1:nitems)
+ {
+ cat <- max(categos[[i]])
+ ifelse(cat==1,names1 <- "remove",names1 <- paste("c",2:cat,sep=""))
+ names2 <- paste("I",i,sep="")
+ names3 <- paste(names1,names2,sep=".")
+ names<- c(names,names3)
+ }
+ names <- names[-1]
+ if(length(grep("remove",names)>0)) names <- names[-grep("remove",names)]
+ colnames(ct.bu) <- names
+ return(ct.bu)
+ }
diff --git a/R/logLik.eRm.r b/R/logLik.eRm.r
old mode 100644
new mode 100755
diff --git a/R/logLik.ppar.r b/R/logLik.ppar.r
old mode 100644
new mode 100755
diff --git a/R/model.matrix.eRm.R b/R/model.matrix.eRm.R
old mode 100644
new mode 100755
diff --git a/R/performance.R b/R/performance.R
old mode 100644
new mode 100755
diff --git a/R/performance_measures.R b/R/performance_measures.R
old mode 100644
new mode 100755
diff --git a/R/performance_plots.R b/R/performance_plots.R
old mode 100644
new mode 100755
diff --git a/R/person.parameter.R b/R/person.parameter.R
old mode 100644
new mode 100755
diff --git a/R/person.parameter.eRm.R b/R/person.parameter.eRm.R
old mode 100644
new mode 100755
diff --git a/R/personfit.R b/R/personfit.R
old mode 100644
new mode 100755
diff --git a/R/personfit.ppar.R b/R/personfit.ppar.R
old mode 100644
new mode 100755
diff --git a/R/pifit.internal.r b/R/pifit.internal.r
old mode 100644
new mode 100755
diff --git a/R/plist.internal.R b/R/plist.internal.R
old mode 100644
new mode 100755
diff --git a/R/plot.ppar.r b/R/plot.ppar.r
old mode 100644
new mode 100755
diff --git a/R/plotCI.R b/R/plotCI.R
old mode 100644
new mode 100755
diff --git a/R/plotDIF.R b/R/plotDIF.R
old mode 100644
new mode 100755
index 7865b95..8e74a8e
--- a/R/plotDIF.R
+++ b/R/plotDIF.R
@@ -1,163 +1,156 @@
plotDIF <- function(object, item.subset=NULL, gamma = 0.95, main=NULL,
- xlim=NULL, xlab=" ", ylab=" ", col=NULL, distance=10,
- splitnames=NULL, leg=FALSE, legpos="bottomleft",...)
-{
-
-if(class(object)=="LR") ## added rh 11-03-17
- object<-list(object)
-else if(is.list(object)) {
- checklr <- sapply(object,class)
- if (!all(checklr=="LR"))
- stop("Elements of '",deparse(substitute(object)), "' must must be LRtest objects!")
-} else if(!is.list(object))
- stop(deparse(substitute(object)), "must be a list of LRtest objects!")
-
-
-# extract number of LRtest objects
-M <- length(sapply(object, function(x) length(x)))
-
-# Confidence plot only for LRtest objects
-for(p in 1:M){
+ xlim=NULL, xlab=" ", ylab=" ", col=NULL, distance,
+ splitnames=NULL, leg=FALSE, legpos="bottomleft", ...){
+
+ if(class(object)=="LR"){ ## added rh 11-03-17
+ object <- list(object)
+ } else if(is.list(object)) {
+ checklr <- sapply(object,class)
+ if(!all(checklr=="LR")) stop("Elements of '",deparse(substitute(object)), "' must must be LRtest objects!")
+ } else if(!is.list(object)) {
+ stop(deparse(substitute(object)), "must be a list of LRtest objects!")
+ }
+
+ # extract number of LRtest objects
+ M <- length(sapply(object, function(x) length(x)))
+
+ # Confidence plot only for LRtest objects
+ for(p in 1:M){
if((object[[p]]$model == "LLTM") || (object[[p]]$model == "LRSM") || (object[[p]]$model == "LPCM")){
- stop("Confidence Plot is computed only for LRtest objects (RM, PCM, RSM)!")
- } else
- if(is.na(sum(unlist(object[[p]]$selist))) == TRUE){
- stop("Confidence Plot is computed only for LRtest objects (RM) with standard errors (se=TRUE)!")
+ stop("Confidence Plot is computed only for LRtest objects (RM, PCM, RSM)!")
+ } else if(is.na(sum(unlist(object[[p]]$selist))) == TRUE){
+ stop("Confidence Plot is computed only for LRtest objects (RM) with standard errors (se=TRUE)!")
+ }
}
-}
-
-
-# confidences list for storing confints
-confidences1 <- vector("list")
-# for labeling list entries
-nam.lab <- vector("character")
-# subgroups splits
-n2 <- sapply(object, function(x) length(x$spl.gr))
-
-# loops for computing thresholds on LRtest objects
- for(m in 1:M){
- # confidences for dichotomous items
- if(object[[m]]$model == "RM"){
- confidences1[[m]] <- lapply(object[[m]]$fitobj,function(x){confint(x, level=gamma)})
- } else {
- # confidences for polytomous items
- confidences1[[m]] <- lapply(object[[m]]$fitobj, function(x){confint(thresholds(x),level=gamma)})
+
+
+ # confidences list for storing confints
+ confidences1 <- vector("list")
+ # for labeling list entries
+ nam.lab <- vector("character")
+ # subgroups splits
+ n2 <- sapply(object, function(x) length(x$spl.gr))
+
+ # loops for computing thresholds on LRtest objects
+ for(m in 1:M){ # confidences for dichotomous items
+ if(object[[m]]$model == "RM"){
+ confidences1[[m]] <- lapply(object[[m]]$fitobj,function(x){confint(x, level=gamma)})
+ } else { # confidences for polytomous items
+ confidences1[[m]] <- lapply(object[[m]]$fitobj, function(x){confint(thresholds(x),level=gamma)})
}
-}
+ }
+ if(is.null(names(object)) == TRUE){
+ names(confidences1) <- paste("LRtest", 1:M, sep="")
+ } else {
+ names(confidences1) <- names(object)
+ }
+ confidences <- do.call(c,lapply(confidences1,function(x) x[1:length(x)]))
-if(is.null(names(object)) == TRUE){
- names(confidences1) <- paste("LRtest", 1:M, sep="")
-} else {
- names(confidences1) <- names(object)
-}
-confidences <- do.call(c,lapply(confidences1,function(x) x[1:length(x)]))
+ if(missing(distance)) distance <- .7/(length(confidences))
+ if((distance <= 0) | (distance >= .5)) stop("distance must not be >= .5 or <= 0")
-model.vec <- vector("character")
-for(p in 1:M){
-model.vec[p] <- object[[p]]$model
-}
- if(any(model.vec == "PCM") || any(model.vec == "RSM")){
- model <- "PCM"
- } else {
- model <- "RM" }
+
+
+ model.vec <- vector("character")
+ for(p in 1:M) model.vec[p] <- object[[p]]$model
-# extracting the longest element of confidences for definition of tickpositions and ticklabels
-# (snatches at the confidences-object index)
-factorlist <- (unique(unlist(lapply(confidences, function(x) dimnames(x)[[1]]))))
-#maxlist <- max(order(factorlist))
-
- if(is.null(item.subset)){
- if(model == "PCM"){
+ if(any(model.vec == "PCM") || any(model.vec == "RSM")){
+ model <- "PCM"
+ } else {
+ model <- "RM"
+ }
+
+ # extracting the longest element of confidences for definition of tickpositions and ticklabels
+ # (snatches at the confidences-object index)
+ factorlist <- (unique(unlist(lapply(confidences, function(x) dimnames(x)[[1]]))))
+ #maxlist <- max(order(factorlist))
+
+ if(is.null(item.subset)){
+ if(model == "PCM"){
y.lab <- sub("thresh beta ", "", factorlist)
- } else {
+ } else {
y.lab <- sub("beta ", "", factorlist)
- }
- } else
- if(is.character(item.subset)){
-# item subset specified as character
- if(model == "PCM"){
- y.lab <- sub("(.+)[.][^.]+$", "\\1", sub("thresh beta ", "", factorlist)) # search only for a "." separation after item label
- categ <- gsub("^.*\\.(.*)$","\\1", factorlist) # extract item categories - search only for a "." separation
- y.lab.id <- y.lab %in% item.subset
- y.lab1 <- y.lab[y.lab.id]
- categ1 <- categ[y.lab.id]
- y.lab <- paste(y.lab1, categ1, sep=".") # stick item categories and names together again
- factorlist <- factorlist[y.lab.id]
- } else {
- y.lab <- sub("beta ", "", factorlist) # search only for a "." separation after item label
- y.lab.id <- y.lab %in% item.subset
- y.lab <- y.lab[y.lab.id]
- factorlist <- factorlist[y.lab.id]
- }
- } else {
-# item subset specified as position number (index in data matrix)
- if(model == "PCM"){
- y.lab <- sub("(.+)[.][^.]+$", "\\1", sub("thresh beta ", "", factorlist)) # search only for a "." separation after item label
- categ <- gsub("^.*\\.(.*)$","\\1", factorlist) # extract item categories - search only for a "." separation
- y.lab2 <- unique(y.lab)[item.subset]
- y.lab.id <- y.lab %in% y.lab2
- y.lab1 <- y.lab[y.lab.id]
- categ1 <- categ[y.lab.id]
- y.lab <- paste(y.lab1, categ1, sep=".") # stick item categories and names together again
- factorlist <- factorlist[y.lab.id]
- } else {
- y.lab <- sub("beta ", "", factorlist)
- y.lab2 <- unique(y.lab)[item.subset]
- y.lab.id <- y.lab %in% y.lab2
- y.lab <- y.lab[y.lab.id]
- factorlist <- factorlist[y.lab.id]
}
-}
-
-# setting range of xaxis
-if(is.null(xlim)==TRUE){
- xlim <- range(unlist(confidences))
-}
-
-# setting tickpositions
-tickpos <- 1:(length(y.lab)) + 3.5/distance
-lty <- unlist(lapply(n2, function(i)1:i))
-
-# defining the plot
-if(is.null(main)) {
- main<-paste("Confidence plot")
-}
-plot(xlim, xlim=xlim, ylim=c(length(y.lab) + 1.25,0.5), type="n", yaxt="n", main=main, xlab=xlab,ylab=ylab,...) # rh 2011-03-23 reverse ylim added
-axis(2, at=tickpos, labels=y.lab, cex.axis=0.7, las=2)
-if(is.null(col)){
-for(k in 1:length(confidences)) {
- for(l in 1:length(factorlist)) {
- lines(as.data.frame(confidences[[k]])[factorlist[l],],
- c(l+k/distance, l+k/distance), t="b", col=length(cumsum(n2)[cumsum(n2) < k])+1, lty=lty[k])}
+ } else if(is.character(item.subset)){ # item subset specified as character
+ if(model == "PCM"){
+ y.lab <- sub("(.+)[.][^.]+$", "\\1", sub("thresh beta ", "", factorlist)) # search only for a "." separation after item label
+ categ <- gsub("^.*\\.(.*)$","\\1", factorlist) # extract item categories - search only for a "." separation
+ y.lab.id <- y.lab %in% item.subset
+ y.lab1 <- y.lab[y.lab.id]
+ categ1 <- categ[y.lab.id]
+ y.lab <- paste(y.lab1, categ1, sep=".") # stick item categories and names together again
+ factorlist <- factorlist[y.lab.id]
+ } else {
+ y.lab <- sub("beta ", "", factorlist) # search only for a "." separation after item label
+ y.lab.id <- y.lab %in% item.subset
+ y.lab <- y.lab[y.lab.id]
+ factorlist <- factorlist[y.lab.id]
+ }
+ } else { # item subset specified as position number (index in data matrix)
+ if(model == "PCM"){
+ y.lab <- sub("(.+)[.][^.]+$", "\\1", sub("thresh beta ", "", factorlist)) # search only for a "." separation after item label
+ categ <- gsub("^.*\\.(.*)$","\\1", factorlist) # extract item categories - search only for a "." separation
+ y.lab2 <- unique(y.lab)[item.subset]
+ y.lab.id <- y.lab %in% y.lab2
+ y.lab1 <- y.lab[y.lab.id]
+ categ1 <- categ[y.lab.id]
+ y.lab <- paste(y.lab1, categ1, sep=".") # stick item categories and names together again
+ factorlist <- factorlist[y.lab.id]
+ } else {
+ y.lab <- sub("beta ", "", factorlist)
+ y.lab2 <- unique(y.lab)[item.subset]
+ y.lab.id <- y.lab %in% y.lab2
+ y.lab <- y.lab[y.lab.id]
+ factorlist <- factorlist[y.lab.id]
+ }
+ }
+
+ # setting range of xaxis
+ if(is.null(xlim)){ xlim <- range(unlist(confidences)) }
+
+ # setting tickpositions
+ tickpos <- 1:(length(y.lab)) # + 3.5/distance mm 2011-06-03
+ lty <- unlist(lapply(n2, function(i)1:i))
+
+ # defining the plot
+ if(is.null(main)){ main<-paste("Confidence plot") }
+
+ plot(xlim, xlim=xlim, ylim=c(1,length(factorlist))+c(-.5,+.5), type="n", yaxt="n", main=main, xlab=xlab,ylab=ylab,...) # rh 2011-03-23 reverse ylim added
+ axis(2, at=tickpos, labels=y.lab, cex.axis=0.7, las=2)
+ if(is.null(col)){
+ for(k in 1:length(confidences)) {
+ for(l in 1:length(factorlist)) {
+ lines(as.data.frame(confidences[[k]])[factorlist[l],],
+ rep(seq(l-.5+distance, l+.5-distance, length.out=length(confidences))[k], 2),
+ type="b", pch=c("[","]"), col=length(cumsum(n2)[cumsum(n2) < k])+1, lty=lty[k])
+ }
+ }
+ } else {
+ col <- rep(col, n2)
+ for(k in 1:length(confidences)) {
+ for(l in 1:length(factorlist)) {
+ lines(as.data.frame(confidences[[k]])[factorlist[l],],
+ rep(seq(l-.5+distance, l+.5-distance, length.out=length(confidences))[k], 2),
+ type="b", pch=c("[","]"), col=col[k], lty=lty[k])
}
-} else {
-col <- rep(col, n2)
-for(k in 1:length(confidences)) {
- for(l in 1:length(factorlist)) {
- lines(as.data.frame(confidences[[k]])[factorlist[l],],
- c(l+k/distance, l+k/distance), t="b", col=col[k], lty=lty[k])}
+ }
}
-}
-
-# doing nicer legend labels
-if(is.null(splitnames)==FALSE){
- names(confidences) <- splitnames
-}
-
-if(leg == TRUE){
- if(is.null(col)){
- ##col <- rep(1:length(n2), each=n2) rh 2011-03-18
- col <- rep(1:length(n2), n2)
-## legend(legpos, rev(paste(names(confidences))), col=rev(col), lty=rev(lty))
- legend(legpos, paste(names(confidences)), col=col, lty=lty)
- } else {
-## legend(legpos, rev(paste(names(confidences))), col=rev(col), lty=rev(lty))
- legend(legpos, paste(names(confidences)), col=col, lty=lty)
- }
-}
-
-invisible(list(confints=confidences1)) #rh 2011-03-18
+
+ # doing nicer legend labels
+ if(is.null(splitnames)==FALSE){ names(confidences) <- splitnames }
+
+ if(leg == TRUE){
+ linespread <- .7 + .3 * (1/length(confidences))
+ if(is.null(col)){ #col <- rep(1:length(n2), each=n2) rh 2011-03-18
+ col <- rep(1:length(n2), n2) # legend(legpos, rev(paste(names(confidences))), col=rev(col), lty=rev(lty))
+ legend(legpos, rev(paste(names(confidences))), y.intersp=linespread, col=rev(col), lty=rev(lty))
+ } else { # legend(legpos, rev(paste(names(confidences))), col=rev(col), lty=rev(lty))
+ legend(legpos, rev(paste(names(confidences))), y.intersp=linespread, col=rev(col), lty=rev(lty))
+ }
+ }
+
+ invisible(list(confints=confidences1)) #rh 2011-03-18
}
diff --git a/R/plotGOF.LR.R b/R/plotGOF.LR.R
old mode 100644
new mode 100755
index 93e28ff..fb3d7a9
--- a/R/plotGOF.LR.R
+++ b/R/plotGOF.LR.R
@@ -8,7 +8,8 @@ function(x,beta.subset="all", main="Graphical Model Check", xlab=NULL,ylab=NULL,
# tlab ... labelling: "item" abbreviated beta parameter name, "number" number from beta par list,
# "identify" interactive, "none"
# pos ... (where the textlabel appears)
-# conf ... confidence ellipses: NULL or list(gamma=0.95, col="red", ia=TRUE, lty="dashed")
+# conf ... confidence ellipses: NULL or
+# list(gamma=0.95, col="red", ia=TRUE, lty="dashed", which=all items in beta.subset)
# ctrline ... control lines (confidence bands): NULL or list(gamma=0.95,lty="solid", col="blue")
# ... additional graphic parameters
@@ -41,6 +42,8 @@ if (is.character(beta.subset)) {
}
} else {
#textlab <- names(beta1)[beta.subset]
+ ##beta.subset<-sort(beta.subset)
+
switch(EXPR=tlab,
item=textlab <- substr(names(beta1)[beta.subset],6,100), #remove "beta " from names
number=textlab <- beta.subset,
@@ -118,12 +121,27 @@ if(is.list(conf)){
col = col)
}
+ # select items for which ellipses are drawn ## rh 2011-05-31
+ if(is.null(conf$which)) conf$which<-beta.subset#seq_along(beta.subset)
+ ##conf$which <- sort(conf$which)
+ if(!all(conf$which %in% beta.subset))
+ stop("Incorrect item number(s) for which ellipses are to be drawn")
+ if(is.null(conf$col)) {
+ conf$c <- rep("red",length.out=length(beta1))
+ } else if (!is.null(conf$which)){
+## conf$c <- rep(NA,length.out=length(beta.subset))
+ conf$c <- rep(NA,length.out=length(conf$which))
+ if (length(conf$c)!=length(conf$which))
+ stop("which and col must have the same length in specification of conf")
+ else
+ conf$c[conf$which]<-conf$col
+ }
+ conf$col <- conf$c
+
if(is.null(conf$gamma)) conf$gamma <- 0.95
- if(is.null(conf$col)) conf$col <- "red"
if(is.null(conf$lty)) conf$lty <- "dotted"
if(is.null(conf$ia)) conf$ia <- FALSE
-
z <- qnorm((conf$gamma+1)/2)
ci1u <- beta1 + z*s1
@@ -148,9 +166,9 @@ if(is.list(conf)){
if(!length(ans)) break
ans <- which(!sel)[ans]
i <- ans
- lines(rep(x[i],2),c(ci2u[i],ci2l[i]),col=conf$col, lty=conf$lty)
- lines(c(ci1u[i],ci1l[i]), rep(y[i],2),col=conf$col,lty=conf$lty)
- ellipse(center=c(x[i],y[i]),matrix(c(v1[i],0,0,v2[i]),2),z,segments=200,center.cex=0.5,lwd=1, col=conf$col)
+ lines(rep(x[i],2),c(ci2u[i],ci2l[i]),col=conf$col[1], lty=conf$lty)
+ lines(c(ci1u[i],ci1l[i]), rep(y[i],2),col=conf$col[1],lty=conf$lty)
+ ellipse(center=c(x[i],y[i]),matrix(c(v1[i],0,0,v2[i]),2),z,segments=200,center.cex=0.5,lwd=1, col=conf$col[1])
#points(x[ans], y[ans], pch = pch)
sel[ans] <- TRUE
res <- c(res, ans)
@@ -164,12 +182,14 @@ if(is.list(conf)){
# non-interactive: plot of all ellipses at once
+ x<-beta1
+ y<-beta2
for (i in beta.subset) {
- x<-beta1
- y<-beta2
- lines(rep(x[i],2),c(ci2u[i],ci2l[i]),col=conf$col, lty=conf$lty)
- lines(c(ci1u[i],ci1l[i]), rep(y[i],2),col=conf$col,lty=conf$lty)
- ellipse(center=c(x[i],y[i]),matrix(c(v1[i],0,0,v2[i]),2),z,segments=200,center.cex=0.5,lwd=1, col=conf$col)
+ if(i %in% conf$which){
+ lines(rep(x[i],2),c(ci2u[i],ci2l[i]),col=conf$col[i], lty=conf$lty)
+ lines(c(ci1u[i],ci1l[i]), rep(y[i],2),col=conf$col[i],lty=conf$lty)
+ ellipse(center=c(x[i],y[i]),matrix(c(v1[i],0,0,v2[i]),2),z,segments=200,center.cex=0.5,lwd=1, col=conf$col[i])
+ }
}
}
}
diff --git a/R/plotGOF.R b/R/plotGOF.R
old mode 100644
new mode 100755
diff --git a/R/plotGR.R b/R/plotGR.R
new file mode 100755
index 0000000..dae9b93
--- /dev/null
+++ b/R/plotGR.R
@@ -0,0 +1,54 @@
+plotGR <- function(object,...)
+ {
+ #TODO: *Add CI around point estimates
+ require(lattice)
+ itms <- object$itms
+ tps <- object$mpoints
+ pplgrps <- object$ngroups/itms
+ if(pplgrps<2) stop("There are no treatment effects in this analysis.")
+
+ #treatment effects for all treatment groups at tps>1
+ treat <- object$etapar[1:((pplgrps-1)*itms*(tps-1))]
+ time <- factor(rep(paste("t",2:tps,sep=""),each=itms*(pplgrps-1)))
+ item <- factor(rep(rep(paste("Item",1:itms),each=pplgrps-1),tps-1))
+ names1 <- unique(names(object$groupvec))[1:(length(unique(names(object$groupvec))))-1]
+ #labeling
+ group <- factor(rep(names1,itms*(tps-1)))
+ plotdats1 <- data.frame(treat,group,item,time)
+
+ #effects (i.e. zeros) for all treatment groups at tp=1
+ treat0 <- rep(0,itms*(pplgrps-1))
+ time0 <- factor(rep("t1",each=itms*(pplgrps-1)))
+ item0 <- factor(rep(paste("Item",1:itms),each=(pplgrps-1)))
+ #labeling
+ group0 <- factor(rep(names1,itms))
+ plotdats0 <- data.frame(treat0,group0,item0,time0)
+ names(plotdats0) <- c("treat","group","item","time")
+
+ #effects (i.e. zeros) for control or baseline group for all tps
+ treat00 <- rep(0,itms*tps)
+ time00 <- factor(rep(paste("t",1:tps,sep=""),each=itms))
+ item00 <- factor(rep(paste("Item",1:itms),tps))
+ group00 <- factor(rep(unique(names(object$groupvec))[length(unique(names(object$groupvec)))],itms*tps))
+ plotdats00 <- data.frame(treat00,group00,item00,time00)
+ names(plotdats00) <- c("treat","group","item","time")
+
+ #all together
+ plotdats <- rbind(plotdats00,plotdats0,plotdats1)
+
+ #plot
+ key.group <- list(space = "right", text = list(levels(plotdats$group)),
+ points = list(pch = 1:length(levels(plotdats$group)),
+ col = "black")
+ )
+ plotout <- xyplot(treat ~ time | item, plotdats,
+ aspect = "xy", type = "o",
+ groups = group, key = key.group,
+ lty = 1, pch = 1:length(levels(plotdats$group)),
+ col.line = "darkgrey", col.symbol = "black",
+ xlab = "Time",
+ ylab = "Effect",
+ main = "Treatment effect plot for LLRA"
+ )
+ print(plotout)
+ }
diff --git a/R/plotICC.R b/R/plotICC.R
old mode 100644
new mode 100755
diff --git a/R/plotICC.Rm.R b/R/plotICC.Rm.R
old mode 100644
new mode 100755
diff --git a/R/plotPImap.R b/R/plotPImap.R
old mode 100644
new mode 100755
diff --git a/R/plotPWmap.R b/R/plotPWmap.R
old mode 100644
new mode 100755
diff --git a/R/plotTR.R b/R/plotTR.R
new file mode 100755
index 0000000..77fe609
--- /dev/null
+++ b/R/plotTR.R
@@ -0,0 +1,31 @@
+
+plotTR <-function(object,...)
+ {
+ #TODO : *Add CI around point estimates
+ require(lattice)
+ #plot trend over time for all items
+ itms <- object$itms
+ tps <- object$mpoints
+ pplgrps <- object$ngroups/itms
+ trend <- object$etapar[((pplgrps-1)*itms*(tps-1)+1):((pplgrps-1)*itms*(tps-1)+(itms*(tps-1)))]
+ tips <-rep(paste("t",1:tps,sep=""),each=itms)
+ items <- rep(paste("Item",1:itms),tps)
+ tr0 <- rep(0,itms)
+ trend <- c(tr0,trend)
+ plotdats <- data.frame(trend,items,tips)
+ key.items <- list(space = "right", text = list(levels(plotdats$items)),
+ points = list(pch = 1:length(levels(plotdats$items)),
+ col = "black")
+ )
+ plotout <- xyplot(trend~tips,data=plotdats,
+ aspect="fill", type="o",
+ groups=items,
+ key=key.items,
+ lty=1,pch = 1:length(levels(plotdats$items)),
+ col.line = "darkgrey", col.symbol = "black",
+ xlab = "Time",
+ ylab = "Effect",
+ main = "Trend effect plot for LLRA"
+ )
+ print(plotout)
+ }
diff --git a/R/plotjointICC.R b/R/plotjointICC.R
old mode 100644
new mode 100755
diff --git a/R/plotjointICC.dRm.R b/R/plotjointICC.dRm.R
old mode 100644
new mode 100755
diff --git a/R/pmat.R b/R/pmat.R
old mode 100644
new mode 100755
diff --git a/R/pmat.default.R b/R/pmat.default.R
old mode 100644
new mode 100755
diff --git a/R/pmat.ppar.R b/R/pmat.ppar.R
old mode 100644
new mode 100755
diff --git a/R/prediction.R b/R/prediction.R
old mode 100644
new mode 100755
diff --git a/R/print.ICr.r b/R/print.ICr.r
old mode 100644
new mode 100755
diff --git a/R/print.LR.R b/R/print.LR.R
old mode 100644
new mode 100755
diff --git a/R/print.MLoef.r b/R/print.MLoef.r
old mode 100644
new mode 100755
index 5deeabe..c2b7e48
--- a/R/print.MLoef.r
+++ b/R/print.MLoef.r
@@ -18,5 +18,5 @@ print.MLoef <- function(x,...)
cat(paste("LR-value:",round(x$LR,3),"\n"))
cat(paste("Chi-square df:",round(x$df,3),"\n"))
cat(paste("p-value:",round(x$p.value,3)),"\n")
- cat("\n")
+ if (!("MLx" %in% class(x))) cat("\n") # no blank line if called from print.MLobj (NPtest)
}
diff --git a/R/print.eRm.R b/R/print.eRm.R
old mode 100644
new mode 100755
diff --git a/R/print.ifit.R b/R/print.ifit.R
old mode 100644
new mode 100755
diff --git a/R/print.llra.R b/R/print.llra.R
new file mode 100755
index 0000000..9dc0a83
--- /dev/null
+++ b/R/print.llra.R
@@ -0,0 +1,10 @@
+print.llra <- function(x,...)
+ {
+ cat("Call:\n")
+ print(x$call)
+ cat("\n")
+ cat("\nParameters:\n")
+ outmat <- rbind(x$etapar,x$se.eta)
+ rownames(outmat) <- c("Estimate","Std.Err")
+ print(outmat)
+ }
diff --git a/R/print.logLik.eRm.R b/R/print.logLik.eRm.R
old mode 100644
new mode 100755
diff --git a/R/print.logLik.ppar.r b/R/print.logLik.ppar.r
old mode 100644
new mode 100755
diff --git a/R/print.pfit.R b/R/print.pfit.R
old mode 100644
new mode 100755
diff --git a/R/print.ppar.R b/R/print.ppar.R
old mode 100644
new mode 100755
diff --git a/R/print.resid.R b/R/print.resid.R
old mode 100644
new mode 100755
diff --git a/R/print.step.r b/R/print.step.r
old mode 100644
new mode 100755
diff --git a/R/print.summary.llra.R b/R/print.summary.llra.R
new file mode 100755
index 0000000..ba88a48
--- /dev/null
+++ b/R/print.summary.llra.R
@@ -0,0 +1,20 @@
+
+print.summary.llra <- function(x,...)
+ {
+ cat("\n")
+ cat("Results of LLRA via",x$model,"estimation: \n")
+ cat("\n")
+ cat("Call: ", x$call, "\n")
+ cat("\n")
+ cat("Conditional log-likelihood:", x$loglik, "\n")
+ cat("Number of iterations:", x$iter, "\n")
+ cat("Number of parameters:", x$npar, "\n")
+ cat("\n")
+ cat("Estimated parameters ")
+ cat("with 0.95 CI:\n")
+ coeftable <- as.data.frame(cbind(round(x$etapar, 3),round(x$se.eta, 3), round(x$ci, 3)))
+ colnames(coeftable) <- c("Estimate", "Std.Error", "lower.CI", "upper.CI")
+ rownames(coeftable) <- names(x$etapar)
+ print(coeftable)
+ cat("\nReference Group: ",x$refGroup,"\n")
+ }
diff --git a/R/print.threshold.r b/R/print.threshold.r
old mode 100644
new mode 100755
diff --git a/R/print.wald.R b/R/print.wald.R
old mode 100644
new mode 100755
diff --git a/R/residuals.ppar.R b/R/residuals.ppar.R
old mode 100644
new mode 100755
diff --git a/R/sim.2pl.R b/R/sim.2pl.R
old mode 100644
new mode 100755
diff --git a/R/sim.locdep.R b/R/sim.locdep.R
old mode 100644
new mode 100755
diff --git a/R/sim.rasch.R b/R/sim.rasch.R
old mode 100644
new mode 100755
diff --git a/R/sim.xdim.R b/R/sim.xdim.R
old mode 100644
new mode 100755
diff --git a/R/stepwiseIt.R b/R/stepwiseIt.R
old mode 100644
new mode 100755
diff --git a/R/stepwiseIt.eRm.R b/R/stepwiseIt.eRm.R
old mode 100644
new mode 100755
diff --git a/R/summary.LR.r b/R/summary.LR.r
old mode 100644
new mode 100755
diff --git a/R/summary.MLoef.r b/R/summary.MLoef.r
old mode 100644
new mode 100755
diff --git a/R/summary.eRm.R b/R/summary.eRm.R
old mode 100644
new mode 100755
diff --git a/R/summary.llra.R b/R/summary.llra.R
new file mode 100755
index 0000000..12c129d
--- /dev/null
+++ b/R/summary.llra.R
@@ -0,0 +1,17 @@
+summary.llra <- function(object, ...) UseMethod("summary.llra")
+
+summary.llra <- function(object, gamma=0.95, ...)
+ #summary for class llra
+ {
+ modi <- object$model
+ calli <- deparse(object$call)
+ logli <- object$loglik
+ iti <- object$iter
+ pari <- object$npar
+ cii <- confint(object, "eta", level=gamma)
+ se.eta <- object$se.eta
+ names(se.eta) <- names(object$etapar)
+ res <- list(etapar=object$etapar,se.eta=se.eta,ci=cii,iter=iti,model=modi,call=calli,npar=pari,loglik=logli,refGroup=object$refGroup)
+ class(res) <- "summary.llra"
+ res
+ }
diff --git a/R/summary.ppar.R b/R/summary.ppar.R
old mode 100644
new mode 100755
diff --git a/R/summary.threshold.r b/R/summary.threshold.r
old mode 100644
new mode 100755
diff --git a/R/thresholds.eRm.r b/R/thresholds.eRm.r
old mode 100644
new mode 100755
diff --git a/R/thresholds.r b/R/thresholds.r
old mode 100644
new mode 100755
diff --git a/R/vcov.eRm.R b/R/vcov.eRm.R
old mode 100644
new mode 100755
diff --git a/R/zzz.R b/R/zzz.R
old mode 100644
new mode 100755
index 9efd0f8..aec2cd9
--- a/R/zzz.R
+++ b/R/zzz.R
@@ -19,7 +19,7 @@ setClass("performance",
y.values = "list",
alpha.values = "list" ))
-setMethod("plot",signature(x="performance",y="missing"),
- function(x,y,...) {
- .plot.performance(x,...)
- })
+#setMethod("plot",signature(x="performance",y="missing"),
+# function(x,y,...) {
+# .plot.performance(x,...)
+# })
diff --git a/data/llraDat1.rda b/data/llraDat1.rda
new file mode 100755
index 0000000..f63e20d
Binary files /dev/null and b/data/llraDat1.rda differ
diff --git a/data/llraDat2.rda b/data/llraDat2.rda
new file mode 100755
index 0000000..58c0762
Binary files /dev/null and b/data/llraDat2.rda differ
diff --git a/data/llradat3.rda b/data/llradat3.rda
new file mode 100755
index 0000000..aa2a06b
Binary files /dev/null and b/data/llradat3.rda differ
diff --git a/data/lltmdat1.rda b/data/lltmdat1.rda
old mode 100644
new mode 100755
index 2d59f46..77bc94c
Binary files a/data/lltmdat1.rda and b/data/lltmdat1.rda differ
diff --git a/data/lltmdat2.rda b/data/lltmdat2.rda
old mode 100644
new mode 100755
index 1ae8982..fd85cd1
Binary files a/data/lltmdat2.rda and b/data/lltmdat2.rda differ
diff --git a/data/lpcmdat.rda b/data/lpcmdat.rda
old mode 100644
new mode 100755
index d5db045..b6876ad
Binary files a/data/lpcmdat.rda and b/data/lpcmdat.rda differ
diff --git a/data/lrsmdat.rda b/data/lrsmdat.rda
old mode 100644
new mode 100755
index cb70680..d28b990
Binary files a/data/lrsmdat.rda and b/data/lrsmdat.rda differ
diff --git a/data/pcmdat.rda b/data/pcmdat.rda
old mode 100644
new mode 100755
index 4c738a3..31c88f5
Binary files a/data/pcmdat.rda and b/data/pcmdat.rda differ
diff --git a/data/pcmdat2.rda b/data/pcmdat2.rda
old mode 100644
new mode 100755
index 325f301..ea9e442
Binary files a/data/pcmdat2.rda and b/data/pcmdat2.rda differ
diff --git a/data/raschdat1.rda b/data/raschdat1.rda
old mode 100644
new mode 100755
index 8875733..f24e872
Binary files a/data/raschdat1.rda and b/data/raschdat1.rda differ
diff --git a/data/raschdat2.rda b/data/raschdat2.rda
old mode 100644
new mode 100755
index 341be27..879c8ae
Binary files a/data/raschdat2.rda and b/data/raschdat2.rda differ
diff --git a/data/rsmdat.rda b/data/rsmdat.rda
old mode 100644
new mode 100755
index 2812985..c62ba4d
Binary files a/data/rsmdat.rda and b/data/rsmdat.rda differ
diff --git a/inst/doc/Rplots.pdf b/inst/doc/Rplots.pdf
deleted file mode 100644
index b554dde..0000000
--- a/inst/doc/Rplots.pdf
+++ /dev/null
@@ -1,4482 +0,0 @@
-%PDF-1.4
-%���ρ�\r
-1 0 obj
-<<
-/CreationDate (D:20101216145553)
-/ModDate (D:20101216145553)
-/Title (R Graphics Output)
-/Producer (R 2.12.0)
-/Creator (R)
->>
-endobj
-2 0 obj
-<<
-/Type /Catalog
-/Pages 3 0 R
->>
-endobj
-5 0 obj
-<<
-/Type /Page
-/Parent 3 0 R
-/Contents 6 0 R
-/Resources 4 0 R
->>
-endobj
-6 0 obj
-<<
-/Length 7 0 R
->>
-stream
-1 J 1 j q
-Q q 59.04 73.44 414.72 371.52 re W n
-0.000 0.000 0.000 RG
-0.75 w
-[] 0 d
-1 J
-1 j
-10.00 M
-BT
-/F1 1 Tf 1 Tr 7.48 0 0 7.48 395.67 378.45 Tm (l) Tj 0 Tr
-ET
-BT
-/F1 1 Tf 1 Tr 7.48 0 0 7.48 318.92 340.90 Tm (l) Tj 0 Tr
-ET
-BT
-/F1 1 Tf 1 Tr 7.48 0 0 7.48 242.83 278.41 Tm (l) Tj 0 Tr
-ET
-BT
-/F1 1 Tf 1 Tr 7.48 0 0 7.48 230.24 203.09 Tm (l) Tj 0 Tr
-ET
-BT
-/F1 1 Tf 1 Tr 7.48 0 0 7.48 165.78 156.44 Tm (l) Tj 0 Tr
-ET
-Q q
-0.000 0.000 0.000 RG
-0.75 w
-[] 0 d
-1 J
-1 j
-10.00 M
-74.40 73.44 m 458.40 73.44 l S
-74.40 73.44 m 74.40 66.24 l S
-138.40 73.44 m 138.40 66.24 l S
-202.40 73.44 m 202.40 66.24 l S
-266.40 73.44 m 266.40 66.24 l S
-330.40 73.44 m 330.40 66.24 l S
-394.40 73.44 m 394.40 66.24 l S
-458.40 73.44 m 458.40 66.24 l S
-BT
-0.000 0.000 0.000 rg
-/F2 1 Tf 12.00 0.00 -0.00 12.00 67.56 47.52 Tm (-3) Tj
-ET
-BT
-/F2 1 Tf 12.00 0.00 -0.00 12.00 131.56 47.52 Tm (-2) Tj
-ET
-BT
-/F2 1 Tf 12.00 0.00 -0.00 12.00 195.56 47.52 Tm (-1) Tj
-ET
-BT
-/F2 1 Tf 12.00 0.00 -0.00 12.00 263.06 47.52 Tm (0) Tj
-ET
-BT
-/F2 1 Tf 12.00 0.00 -0.00 12.00 327.06 47.52 Tm (1) Tj
-ET
-BT
-/F2 1 Tf 12.00 0.00 -0.00 12.00 391.06 47.52 Tm (2) Tj
-ET
-BT
-/F2 1 Tf 12.00 0.00 -0.00 12.00 455.06 47.52 Tm (3) Tj
-ET
-59.04 87.20 m 59.04 431.20 l S
-59.04 87.20 m 51.84 87.20 l S
-59.04 144.53 m 51.84 144.53 l S
-59.04 201.87 m 51.84 201.87 l S
-59.04 259.20 m 51.84 259.20 l S
-59.04 316.53 m 51.84 316.53 l S
-59.04 373.87 m 51.84 373.87 l S
-59.04 431.20 m 51.84 431.20 l S
-BT
-/F2 1 Tf 0.00 12.00 -12.00 0.00 41.76 80.36 Tm (-3) Tj
-ET
-BT
-/F2 1 Tf 0.00 12.00 -12.00 0.00 41.76 137.69 Tm (-2) Tj
-ET
-BT
-/F2 1 Tf 0.00 12.00 -12.00 0.00 41.76 195.03 Tm (-1) Tj
-ET
-BT
-/F2 1 Tf 0.00 12.00 -12.00 0.00 41.76 255.86 Tm (0) Tj
-ET
-BT
-/F2 1 Tf 0.00 12.00 -12.00 0.00 41.76 313.20 Tm (1) Tj
-ET
-BT
-/F2 1 Tf 0.00 12.00 -12.00 0.00 41.76 370.53 Tm (2) Tj
-ET
-BT
-/F2 1 Tf 0.00 12.00 -12.00 0.00 41.76 427.86 Tm (3) Tj
-ET
-59.04 73.44 m
-473.76 73.44 l
-473.76 444.96 l
-59.04 444.96 l
-59.04 73.44 l
-S
-Q q
-BT
-0.000 0.000 0.000 rg
-/F3 1 Tf 14.00 0.00 -0.00 14.00 189.13 469.45 Tm [(Graphical Model Chec) 20 (k)] TJ
-ET
-BT
-/F2 1 Tf 12.00 0.00 -0.00 12.00 169.82 18.72 Tm [(Beta f) 30 (or Group: Ra) 20 (w Scores < Mean)] TJ
-ET
-BT
-/F2 1 Tf 0.00 12.00 -12.00 0.00 12.96 159.12 Tm [(Beta f) 30 (or Group: Ra) 20 (w Scores >= Mean)] TJ
-ET
-Q q 59.04 73.44 414.72 371.52 re W n
-0.000 0.000 0.000 RG
-0.75 w
-[] 0 d
-1 J
-1 j
-10.00 M
-59.04 73.44 m 473.76 444.96 l S
-BT
-0.000 0.000 0.000 rg
-/F2 1 Tf 12.00 0.00 -0.00 12.00 405.83 378.17 Tm (I14) Tj
-ET
-BT
-/F2 1 Tf 12.00 0.00 -0.00 12.00 329.08 340.62 Tm (I5) Tj
-ET
-BT
-/F2 1 Tf 12.00 0.00 -0.00 12.00 253.00 278.13 Tm (I18) Tj
-ET
-BT
-/F2 1 Tf 12.00 0.00 -0.00 12.00 240.40 202.81 Tm (I7) Tj
-ET
-BT
-/F2 1 Tf 12.00 0.00 -0.00 12.00 175.94 156.16 Tm (I1) Tj
-ET
-0.000 0.000 1.000 RG
-0.75 w
-[ 0.00 3.00] 0 d
-398.63 420.80 m
-398.63 341.29 l
-S
-486.83 381.04 m
-310.43 381.04 l
-S
-0.75 w
-[] 0 d
-486.83 381.04 m
-486.78 382.29 l
-486.65 383.54 l
-486.44 384.79 l
-486.13 386.03 l
-485.74 387.26 l
-485.27 388.49 l
-484.70 389.72 l
-484.06 390.93 l
-483.33 392.13 l
-482.51 393.33 l
-481.61 394.51 l
-480.63 395.68 l
-479.57 396.83 l
-478.43 397.97 l
-477.21 399.09 l
-475.92 400.19 l
-474.55 401.28 l
-473.10 402.34 l
-471.58 403.39 l
-469.98 404.41 l
-468.32 405.41 l
-466.59 406.38 l
-464.79 407.33 l
-462.92 408.26 l
-461.00 409.15 l
-459.01 410.02 l
-456.96 410.86 l
-454.85 411.67 l
-452.69 412.45 l
-450.47 413.20 l
-448.20 413.92 l
-445.89 414.61 l
-443.53 415.26 l
-441.12 415.88 l
-438.67 416.46 l
-436.18 417.01 l
-433.66 417.53 l
-431.10 418.00 l
-428.51 418.45 l
-425.89 418.85 l
-423.24 419.22 l
-420.56 419.55 l
-417.87 419.84 l
-415.16 420.09 l
-412.43 420.31 l
-409.68 420.48 l
-406.93 420.62 l
-404.17 420.72 l
-401.40 420.78 l
-398.63 420.80 l
-395.86 420.78 l
-393.09 420.72 l
-390.33 420.62 l
-387.58 420.48 l
-384.83 420.31 l
-382.10 420.09 l
-379.39 419.84 l
-376.70 419.55 l
-374.02 419.22 l
-371.38 418.85 l
-368.76 418.45 l
-366.16 418.00 l
-363.60 417.53 l
-361.08 417.01 l
-358.59 416.46 l
-356.14 415.88 l
-353.74 415.26 l
-351.37 414.61 l
-349.06 413.92 l
-346.79 413.20 l
-344.57 412.45 l
-342.41 411.67 l
-340.31 410.86 l
-338.26 410.02 l
-336.27 409.15 l
-334.34 408.26 l
-332.47 407.33 l
-330.67 406.38 l
-328.94 405.41 l
-327.28 404.41 l
-325.69 403.39 l
-324.16 402.34 l
-322.72 401.28 l
-321.34 400.19 l
-320.05 399.09 l
-318.83 397.97 l
-317.69 396.83 l
-316.63 395.68 l
-315.65 394.51 l
-314.75 393.33 l
-313.94 392.13 l
-313.21 390.93 l
-312.56 389.72 l
-312.00 388.49 l
-311.52 387.26 l
-311.13 386.03 l
-310.83 384.79 l
-310.61 383.54 l
-310.48 382.29 l
-310.43 381.04 l
-310.48 379.80 l
-310.61 378.55 l
-310.83 377.30 l
-311.13 376.06 l
-311.52 374.83 l
-312.00 373.60 l
-312.56 372.37 l
-313.21 371.16 l
-313.94 369.95 l
-314.75 368.76 l
-315.65 367.58 l
-316.63 366.41 l
-317.69 365.26 l
-318.83 364.12 l
-320.05 363.00 l
-321.34 361.89 l
-322.72 360.81 l
-324.16 359.74 l
-325.69 358.70 l
-327.28 357.68 l
-328.94 356.68 l
-330.67 355.71 l
-332.47 354.76 l
-334.34 353.83 l
-336.27 352.94 l
-338.26 352.07 l
-340.31 351.23 l
-342.41 350.42 l
-344.57 349.63 l
-346.79 348.89 l
-349.06 348.17 l
-351.37 347.48 l
-353.74 346.83 l
-356.14 346.21 l
-358.59 345.63 l
-361.08 345.08 l
-363.60 344.56 l
-366.16 344.08 l
-368.76 343.64 l
-371.38 343.24 l
-374.02 342.87 l
-376.70 342.54 l
-379.39 342.25 l
-382.10 342.00 l
-384.83 341.78 l
-387.58 341.61 l
-390.33 341.47 l
-393.09 341.37 l
-395.86 341.31 l
-398.63 341.29 l
-401.40 341.31 l
-404.17 341.37 l
-406.93 341.47 l
-409.68 341.61 l
-412.43 341.78 l
-415.16 342.00 l
-417.87 342.25 l
-420.56 342.54 l
-423.24 342.87 l
-425.89 343.24 l
-428.51 343.64 l
-431.10 344.08 l
-433.66 344.56 l
-436.18 345.08 l
-438.67 345.63 l
-441.12 346.21 l
-443.53 346.83 l
-445.89 347.48 l
-448.20 348.17 l
-450.47 348.89 l
-452.69 349.63 l
-454.85 350.42 l
-456.96 351.23 l
-459.01 352.07 l
-461.00 352.94 l
-462.92 353.83 l
-464.79 354.76 l
-466.59 355.71 l
-468.32 356.68 l
-469.98 357.68 l
-471.58 358.70 l
-473.10 359.74 l
-474.55 360.81 l
-475.92 361.89 l
-477.21 363.00 l
-478.43 364.12 l
-479.57 365.26 l
-480.63 366.41 l
-481.61 367.58 l
-482.51 368.76 l
-483.33 369.95 l
-484.06 371.16 l
-484.70 372.37 l
-485.27 373.60 l
-485.74 374.83 l
-486.13 376.06 l
-486.44 377.30 l
-486.65 378.55 l
-486.78 379.80 l
-486.83 381.04 l
-S
-0.000 0.000 1.000 rg
-BT
-/F1 1 Tf 2 Tr 3.74 0 0 3.74 397.15 379.75 Tm (l) Tj 0 Tr
-ET
-0.75 w
-[ 0.00 3.00] 0 d
-321.88 377.69 m
-321.88 309.29 l
-S
-375.93 343.49 m
-267.82 343.49 l
-S
-0.75 w
-[] 0 d
-375.93 343.49 m
-375.91 344.57 l
-375.83 345.64 l
-375.69 346.71 l
-375.51 347.78 l
-375.27 348.84 l
-374.98 349.90 l
-374.63 350.95 l
-374.24 352.00 l
-373.79 353.03 l
-373.29 354.06 l
-372.74 355.08 l
-372.14 356.08 l
-371.49 357.07 l
-370.79 358.05 l
-370.04 359.02 l
-369.25 359.97 l
-368.41 360.90 l
-367.52 361.82 l
-366.59 362.71 l
-365.61 363.59 l
-364.59 364.45 l
-363.53 365.29 l
-362.43 366.11 l
-361.28 366.90 l
-360.10 367.67 l
-358.88 368.42 l
-357.63 369.14 l
-356.34 369.84 l
-355.01 370.51 l
-353.65 371.16 l
-352.26 371.78 l
-350.84 372.37 l
-349.40 372.93 l
-347.92 373.46 l
-346.42 373.96 l
-344.90 374.44 l
-343.35 374.88 l
-341.78 375.29 l
-340.19 375.67 l
-338.58 376.02 l
-336.96 376.33 l
-335.32 376.62 l
-333.67 376.87 l
-332.01 377.08 l
-330.34 377.27 l
-328.65 377.42 l
-326.97 377.54 l
-325.27 377.62 l
-323.58 377.67 l
-321.88 377.69 l
-320.18 377.67 l
-318.49 377.62 l
-316.79 377.54 l
-315.10 377.42 l
-313.42 377.27 l
-311.75 377.08 l
-310.09 376.87 l
-308.44 376.62 l
-306.80 376.33 l
-305.18 376.02 l
-303.57 375.67 l
-301.98 375.29 l
-300.41 374.88 l
-298.86 374.44 l
-297.34 373.96 l
-295.84 373.46 l
-294.36 372.93 l
-292.92 372.37 l
-291.50 371.78 l
-290.11 371.16 l
-288.75 370.51 l
-287.42 369.84 l
-286.13 369.14 l
-284.88 368.42 l
-283.66 367.67 l
-282.48 366.90 l
-281.33 366.11 l
-280.23 365.29 l
-279.17 364.45 l
-278.15 363.59 l
-277.17 362.71 l
-276.24 361.82 l
-275.35 360.90 l
-274.51 359.97 l
-273.72 359.02 l
-272.97 358.05 l
-272.27 357.07 l
-271.62 356.08 l
-271.02 355.08 l
-270.47 354.06 l
-269.97 353.03 l
-269.52 352.00 l
-269.13 350.95 l
-268.78 349.90 l
-268.49 348.84 l
-268.25 347.78 l
-268.06 346.71 l
-267.93 345.64 l
-267.85 344.57 l
-267.82 343.49 l
-267.85 342.42 l
-267.93 341.34 l
-268.06 340.27 l
-268.25 339.21 l
-268.49 338.14 l
-268.78 337.08 l
-269.13 336.03 l
-269.52 334.99 l
-269.97 333.95 l
-270.47 332.92 l
-271.02 331.91 l
-271.62 330.90 l
-272.27 329.91 l
-272.97 328.93 l
-273.72 327.97 l
-274.51 327.02 l
-275.35 326.08 l
-276.24 325.17 l
-277.17 324.27 l
-278.15 323.39 l
-279.17 322.53 l
-280.23 321.69 l
-281.33 320.88 l
-282.48 320.08 l
-283.66 319.31 l
-284.88 318.56 l
-286.13 317.84 l
-287.42 317.14 l
-288.75 316.47 l
-290.11 315.82 l
-291.50 315.21 l
-292.92 314.62 l
-294.36 314.06 l
-295.84 313.52 l
-297.34 313.02 l
-298.86 312.55 l
-300.41 312.11 l
-301.98 311.70 l
-303.57 311.32 l
-305.18 310.97 l
-306.80 310.65 l
-308.44 310.37 l
-310.09 310.12 l
-311.75 309.90 l
-313.42 309.71 l
-315.10 309.56 l
-316.79 309.45 l
-318.49 309.36 l
-320.18 309.31 l
-321.88 309.29 l
-323.58 309.31 l
-325.27 309.36 l
-326.97 309.45 l
-328.65 309.56 l
-330.34 309.71 l
-332.01 309.90 l
-333.67 310.12 l
-335.32 310.37 l
-336.96 310.65 l
-338.58 310.97 l
-340.19 311.32 l
-341.78 311.70 l
-343.35 312.11 l
-344.90 312.55 l
-346.42 313.02 l
-347.92 313.52 l
-349.40 314.06 l
-350.84 314.62 l
-352.26 315.21 l
-353.65 315.82 l
-355.01 316.47 l
-356.34 317.14 l
-357.63 317.84 l
-358.88 318.56 l
-360.10 319.31 l
-361.28 320.08 l
-362.43 320.88 l
-363.53 321.69 l
-364.59 322.53 l
-365.61 323.39 l
-366.59 324.27 l
-367.52 325.17 l
-368.41 326.08 l
-369.25 327.02 l
-370.04 327.97 l
-370.79 328.93 l
-371.49 329.91 l
-372.14 330.90 l
-372.74 331.91 l
-373.29 332.92 l
-373.79 333.95 l
-374.24 334.99 l
-374.63 336.03 l
-374.98 337.08 l
-375.27 338.14 l
-375.51 339.21 l
-375.69 340.27 l
-375.83 341.34 l
-375.91 342.42 l
-375.93 343.49 l
-S
-BT
-/F1 1 Tf 2 Tr 3.74 0 0 3.74 320.40 342.19 Tm (l) Tj 0 Tr
-ET
-0.75 w
-[ 0.00 3.00] 0 d
-245.80 312.43 m
-245.80 249.57 l
-S
-285.44 281.00 m
-206.15 281.00 l
-S
-0.75 w
-[] 0 d
-285.44 281.00 m
-285.42 281.99 l
-285.36 282.98 l
-285.26 283.96 l
-285.13 284.94 l
-284.95 285.92 l
-284.74 286.89 l
-284.49 287.86 l
-284.19 288.82 l
-283.87 289.77 l
-283.50 290.72 l
-283.10 291.65 l
-282.66 292.57 l
-282.18 293.49 l
-281.67 294.39 l
-281.12 295.27 l
-280.54 296.15 l
-279.92 297.00 l
-279.27 297.85 l
-278.59 298.67 l
-277.87 299.48 l
-277.12 300.27 l
-276.34 301.04 l
-275.53 301.79 l
-274.70 302.52 l
-273.83 303.23 l
-272.93 303.92 l
-272.01 304.58 l
-271.07 305.22 l
-270.09 305.84 l
-269.10 306.43 l
-268.08 307.00 l
-267.04 307.54 l
-265.98 308.06 l
-264.90 308.55 l
-263.79 309.01 l
-262.68 309.44 l
-261.54 309.85 l
-260.39 310.23 l
-259.23 310.58 l
-258.05 310.90 l
-256.86 311.19 l
-255.66 311.45 l
-254.44 311.68 l
-253.23 311.88 l
-252.00 312.05 l
-250.77 312.19 l
-249.53 312.30 l
-248.29 312.37 l
-247.04 312.42 l
-245.80 312.43 l
-244.55 312.42 l
-243.31 312.37 l
-242.07 312.30 l
-240.83 312.19 l
-239.59 312.05 l
-238.37 311.88 l
-237.15 311.68 l
-235.94 311.45 l
-234.74 311.19 l
-233.55 310.90 l
-232.37 310.58 l
-231.20 310.23 l
-230.05 309.85 l
-228.92 309.44 l
-227.80 309.01 l
-226.70 308.55 l
-225.62 308.06 l
-224.55 307.54 l
-223.51 307.00 l
-222.49 306.43 l
-221.50 305.84 l
-220.53 305.22 l
-219.58 304.58 l
-218.66 303.92 l
-217.76 303.23 l
-216.90 302.52 l
-216.06 301.79 l
-215.25 301.04 l
-214.47 300.27 l
-213.72 299.48 l
-213.01 298.67 l
-212.32 297.85 l
-211.67 297.00 l
-211.06 296.15 l
-210.47 295.27 l
-209.93 294.39 l
-209.41 293.49 l
-208.94 292.57 l
-208.50 291.65 l
-208.09 290.72 l
-207.73 289.77 l
-207.40 288.82 l
-207.11 287.86 l
-206.85 286.89 l
-206.64 285.92 l
-206.47 284.94 l
-206.33 283.96 l
-206.23 282.98 l
-206.17 281.99 l
-206.15 281.00 l
-206.17 280.02 l
-206.23 279.03 l
-206.33 278.04 l
-206.47 277.06 l
-206.64 276.09 l
-206.85 275.11 l
-207.11 274.15 l
-207.40 273.19 l
-207.73 272.23 l
-208.09 271.29 l
-208.50 270.36 l
-208.94 269.43 l
-209.41 268.52 l
-209.93 267.62 l
-210.47 266.73 l
-211.06 265.86 l
-211.67 265.00 l
-212.32 264.16 l
-213.01 263.34 l
-213.72 262.53 l
-214.47 261.74 l
-215.25 260.97 l
-216.06 260.22 l
-216.90 259.49 l
-217.76 258.78 l
-218.66 258.09 l
-219.58 257.43 l
-220.53 256.78 l
-221.50 256.17 l
-222.49 255.57 l
-223.51 255.01 l
-224.55 254.46 l
-225.62 253.95 l
-226.70 253.46 l
-227.80 253.00 l
-228.92 252.56 l
-230.05 252.16 l
-231.20 251.78 l
-232.37 251.43 l
-233.55 251.11 l
-234.74 250.82 l
-235.94 250.56 l
-237.15 250.33 l
-238.37 250.13 l
-239.59 249.96 l
-240.83 249.82 l
-242.07 249.71 l
-243.31 249.63 l
-244.55 249.59 l
-245.80 249.57 l
-247.04 249.59 l
-248.29 249.63 l
-249.53 249.71 l
-250.77 249.82 l
-252.00 249.96 l
-253.23 250.13 l
-254.44 250.33 l
-255.66 250.56 l
-256.86 250.82 l
-258.05 251.11 l
-259.23 251.43 l
-260.39 251.78 l
-261.54 252.16 l
-262.68 252.56 l
-263.79 253.00 l
-264.90 253.46 l
-265.98 253.95 l
-267.04 254.46 l
-268.08 255.01 l
-269.10 255.57 l
-270.09 256.17 l
-271.07 256.78 l
-272.01 257.43 l
-272.93 258.09 l
-273.83 258.78 l
-274.70 259.49 l
-275.53 260.22 l
-276.34 260.97 l
-277.12 261.74 l
-277.87 262.53 l
-278.59 263.34 l
-279.27 264.16 l
-279.92 265.00 l
-280.54 265.86 l
-281.12 266.73 l
-281.67 267.62 l
-282.18 268.52 l
-282.66 269.43 l
-283.10 270.36 l
-283.50 271.29 l
-283.87 272.23 l
-284.19 273.19 l
-284.49 274.15 l
-284.74 275.11 l
-284.95 276.09 l
-285.13 277.06 l
-285.26 278.04 l
-285.36 279.03 l
-285.42 280.02 l
-285.44 281.00 l
-S
-BT
-/F1 1 Tf 2 Tr 3.74 0 0 3.74 244.32 279.71 Tm (l) Tj 0 Tr
-ET
-0.75 w
-[ 0.00 3.00] 0 d
-233.20 244.74 m
-233.20 166.63 l
-S
-271.83 205.68 m
-194.58 205.68 l
-S
-0.75 w
-[] 0 d
-271.83 205.68 m
-271.81 206.91 l
-271.75 208.14 l
-271.66 209.36 l
-271.52 210.58 l
-271.35 211.79 l
-271.14 213.00 l
-270.90 214.20 l
-270.62 215.40 l
-270.30 216.58 l
-269.94 217.75 l
-269.55 218.91 l
-269.12 220.06 l
-268.65 221.19 l
-268.15 222.31 l
-267.62 223.41 l
-267.05 224.50 l
-266.45 225.56 l
-265.82 226.61 l
-265.15 227.64 l
-264.45 228.64 l
-263.72 229.62 l
-262.97 230.58 l
-262.18 231.51 l
-261.36 232.42 l
-260.52 233.30 l
-259.64 234.15 l
-258.75 234.98 l
-257.82 235.78 l
-256.88 236.54 l
-255.91 237.28 l
-254.91 237.99 l
-253.90 238.66 l
-252.87 239.30 l
-251.81 239.91 l
-250.74 240.48 l
-249.65 241.02 l
-248.54 241.53 l
-247.42 242.00 l
-246.29 242.43 l
-245.14 242.83 l
-243.98 243.19 l
-242.81 243.51 l
-241.63 243.80 l
-240.44 244.05 l
-239.25 244.26 l
-238.05 244.43 l
-236.84 244.57 l
-235.63 244.66 l
-234.42 244.72 l
-233.20 244.74 l
-231.99 244.72 l
-230.78 244.66 l
-229.57 244.57 l
-228.36 244.43 l
-227.16 244.26 l
-225.97 244.05 l
-224.78 243.80 l
-223.60 243.51 l
-222.43 243.19 l
-221.27 242.83 l
-220.12 242.43 l
-218.99 242.00 l
-217.87 241.53 l
-216.76 241.02 l
-215.67 240.48 l
-214.60 239.91 l
-213.54 239.30 l
-212.51 238.66 l
-211.49 237.99 l
-210.50 237.28 l
-209.53 236.54 l
-208.58 235.78 l
-207.66 234.98 l
-206.76 234.15 l
-205.89 233.30 l
-205.05 232.42 l
-204.23 231.51 l
-203.44 230.58 l
-202.69 229.62 l
-201.96 228.64 l
-201.26 227.64 l
-200.59 226.61 l
-199.96 225.56 l
-199.36 224.50 l
-198.79 223.41 l
-198.26 222.31 l
-197.76 221.19 l
-197.29 220.06 l
-196.86 218.91 l
-196.47 217.75 l
-196.11 216.58 l
-195.79 215.40 l
-195.51 214.20 l
-195.26 213.00 l
-195.06 211.79 l
-194.89 210.58 l
-194.75 209.36 l
-194.66 208.14 l
-194.60 206.91 l
-194.58 205.68 l
-194.60 204.46 l
-194.66 203.23 l
-194.75 202.01 l
-194.89 200.79 l
-195.06 199.57 l
-195.26 198.37 l
-195.51 197.16 l
-195.79 195.97 l
-196.11 194.79 l
-196.47 193.62 l
-196.86 192.45 l
-197.29 191.31 l
-197.76 190.17 l
-198.26 189.06 l
-198.79 187.95 l
-199.36 186.87 l
-199.96 185.80 l
-200.59 184.76 l
-201.26 183.73 l
-201.96 182.73 l
-202.69 181.75 l
-203.44 180.79 l
-204.23 179.86 l
-205.05 178.95 l
-205.89 178.07 l
-206.76 177.21 l
-207.66 176.39 l
-208.58 175.59 l
-209.53 174.82 l
-210.50 174.09 l
-211.49 173.38 l
-212.51 172.71 l
-213.54 172.07 l
-214.60 171.46 l
-215.67 170.89 l
-216.76 170.35 l
-217.87 169.84 l
-218.99 169.37 l
-220.12 168.94 l
-221.27 168.54 l
-222.43 168.18 l
-223.60 167.86 l
-224.78 167.57 l
-225.97 167.32 l
-227.16 167.11 l
-228.36 166.94 l
-229.57 166.80 l
-230.78 166.71 l
-231.99 166.65 l
-233.20 166.63 l
-234.42 166.65 l
-235.63 166.71 l
-236.84 166.80 l
-238.05 166.94 l
-239.25 167.11 l
-240.44 167.32 l
-241.63 167.57 l
-242.81 167.86 l
-243.98 168.18 l
-245.14 168.54 l
-246.29 168.94 l
-247.42 169.37 l
-248.54 169.84 l
-249.65 170.35 l
-250.74 170.89 l
-251.81 171.46 l
-252.87 172.07 l
-253.90 172.71 l
-254.91 173.38 l
-255.91 174.09 l
-256.88 174.82 l
-257.82 175.59 l
-258.75 176.39 l
-259.64 177.21 l
-260.52 178.07 l
-261.36 178.95 l
-262.18 179.86 l
-262.97 180.79 l
-263.72 181.75 l
-264.45 182.73 l
-265.15 183.73 l
-265.82 184.76 l
-266.45 185.80 l
-267.05 186.87 l
-267.62 187.95 l
-268.15 189.06 l
-268.65 190.17 l
-269.12 191.31 l
-269.55 192.45 l
-269.94 193.62 l
-270.30 194.79 l
-270.62 195.97 l
-270.90 197.16 l
-271.14 198.37 l
-271.35 199.57 l
-271.52 200.79 l
-271.66 202.01 l
-271.75 203.23 l
-271.81 204.46 l
-271.83 205.68 l
-S
-BT
-/F1 1 Tf 2 Tr 3.74 0 0 3.74 231.72 204.39 Tm (l) Tj 0 Tr
-ET
-0.75 w
-[ 0.00 3.00] 0 d
-168.74 210.70 m
-168.74 107.36 l
-S
-207.11 159.03 m
-130.38 159.03 l
-S
-0.75 w
-[] 0 d
-207.11 159.03 m
-207.09 160.65 l
-207.04 162.28 l
-206.94 163.89 l
-206.81 165.51 l
-206.64 167.11 l
-206.43 168.71 l
-206.19 170.30 l
-205.91 171.88 l
-205.59 173.45 l
-205.23 175.00 l
-204.84 176.53 l
-204.42 178.05 l
-203.96 179.55 l
-203.46 181.03 l
-202.93 182.49 l
-202.37 183.92 l
-201.77 185.33 l
-201.14 186.72 l
-200.48 188.07 l
-199.78 189.40 l
-199.06 190.70 l
-198.31 191.97 l
-197.52 193.20 l
-196.71 194.40 l
-195.87 195.57 l
-195.01 196.70 l
-194.12 197.79 l
-193.20 198.84 l
-192.26 199.86 l
-191.30 200.83 l
-190.31 201.77 l
-189.30 202.66 l
-188.27 203.51 l
-187.23 204.31 l
-186.16 205.07 l
-185.08 205.78 l
-183.98 206.45 l
-182.87 207.07 l
-181.74 207.65 l
-180.60 208.17 l
-179.45 208.65 l
-178.28 209.08 l
-177.11 209.46 l
-175.93 209.79 l
-174.75 210.07 l
-173.55 210.29 l
-172.35 210.47 l
-171.15 210.60 l
-169.95 210.68 l
-168.74 210.70 l
-167.54 210.68 l
-166.33 210.60 l
-165.13 210.47 l
-163.93 210.29 l
-162.74 210.07 l
-161.55 209.79 l
-160.37 209.46 l
-159.20 209.08 l
-158.04 208.65 l
-156.89 208.17 l
-155.75 207.65 l
-154.62 207.07 l
-153.51 206.45 l
-152.41 205.78 l
-151.32 205.07 l
-150.26 204.31 l
-149.21 203.51 l
-148.18 202.66 l
-147.18 201.77 l
-146.19 200.83 l
-145.23 199.86 l
-144.29 198.84 l
-143.37 197.79 l
-142.48 196.70 l
-141.61 195.57 l
-140.77 194.40 l
-139.96 193.20 l
-139.18 191.97 l
-138.43 190.70 l
-137.70 189.40 l
-137.01 188.07 l
-136.35 186.72 l
-135.72 185.33 l
-135.12 183.92 l
-134.56 182.49 l
-134.03 181.03 l
-133.53 179.55 l
-133.07 178.05 l
-132.64 176.53 l
-132.25 175.00 l
-131.90 173.45 l
-131.58 171.88 l
-131.30 170.30 l
-131.05 168.71 l
-130.85 167.11 l
-130.68 165.51 l
-130.55 163.89 l
-130.45 162.28 l
-130.39 160.65 l
-130.38 159.03 l
-130.39 157.41 l
-130.45 155.79 l
-130.55 154.17 l
-130.68 152.55 l
-130.85 150.95 l
-131.05 149.35 l
-131.30 147.76 l
-131.58 146.18 l
-131.90 144.62 l
-132.25 143.06 l
-132.64 141.53 l
-133.07 140.01 l
-133.53 138.51 l
-134.03 137.03 l
-134.56 135.57 l
-135.12 134.14 l
-135.72 132.73 l
-136.35 131.34 l
-137.01 129.99 l
-137.70 128.66 l
-138.43 127.36 l
-139.18 126.09 l
-139.96 124.86 l
-140.77 123.66 l
-141.61 122.49 l
-142.48 121.36 l
-143.37 120.27 l
-144.29 119.22 l
-145.23 118.20 l
-146.19 117.23 l
-147.18 116.29 l
-148.18 115.40 l
-149.21 114.56 l
-150.26 113.75 l
-151.32 112.99 l
-152.41 112.28 l
-153.51 111.61 l
-154.62 110.99 l
-155.75 110.41 l
-156.89 109.89 l
-158.04 109.41 l
-159.20 108.98 l
-160.37 108.60 l
-161.55 108.27 l
-162.74 108.00 l
-163.93 107.77 l
-165.13 107.59 l
-166.33 107.46 l
-167.54 107.39 l
-168.74 107.36 l
-169.95 107.39 l
-171.15 107.46 l
-172.35 107.59 l
-173.55 107.77 l
-174.75 108.00 l
-175.93 108.27 l
-177.11 108.60 l
-178.28 108.98 l
-179.45 109.41 l
-180.60 109.89 l
-181.74 110.41 l
-182.87 110.99 l
-183.98 111.61 l
-185.08 112.28 l
-186.16 112.99 l
-187.23 113.75 l
-188.27 114.56 l
-189.30 115.40 l
-190.31 116.29 l
-191.30 117.23 l
-192.26 118.20 l
-193.20 119.22 l
-194.12 120.27 l
-195.01 121.36 l
-195.87 122.49 l
-196.71 123.66 l
-197.52 124.86 l
-198.31 126.09 l
-199.06 127.36 l
-199.78 128.66 l
-200.48 129.99 l
-201.14 131.34 l
-201.77 132.73 l
-202.37 134.14 l
-202.93 135.57 l
-203.46 137.03 l
-203.96 138.51 l
-204.42 140.01 l
-204.84 141.53 l
-205.23 143.06 l
-205.59 144.62 l
-205.91 146.18 l
-206.19 147.76 l
-206.43 149.35 l
-206.64 150.95 l
-206.81 152.55 l
-206.94 154.17 l
-207.04 155.79 l
-207.09 157.41 l
-207.11 159.03 l
-S
-BT
-/F1 1 Tf 2 Tr 3.74 0 0 3.74 167.26 157.73 Tm (l) Tj 0 Tr
-ET
-Q
-endstream
-endobj
-7 0 obj
-19696
-endobj
-8 0 obj
-<<
-/Type /Page
-/Parent 3 0 R
-/Contents 9 0 R
-/Resources 4 0 R
->>
-endobj
-9 0 obj
-<<
-/Length 10 0 R
->>
-stream
-1 J 1 j q
-Q q 49.00 312.96 177.90 142.04 re W n
-Q q 49.00 312.96 177.90 142.04 re W n
-0.000 0.000 0.000 RG
-0.75 w
-[] 0 d
-1 J
-1 j
-10.00 M
-55.59 448.42 m
-57.65 448.28 l
-59.71 448.13 l
-61.77 447.97 l
-63.83 447.78 l
-65.89 447.58 l
-67.95 447.36 l
-70.00 447.11 l
-72.06 446.84 l
-74.12 446.54 l
-76.18 446.22 l
-78.24 445.85 l
-80.30 445.46 l
-82.36 445.02 l
-84.42 444.54 l
-86.48 444.02 l
-88.54 443.44 l
-90.59 442.81 l
-92.65 442.12 l
-94.71 441.36 l
-96.77 440.53 l
-98.83 439.63 l
-100.89 438.65 l
-102.95 437.57 l
-105.01 436.41 l
-107.07 435.14 l
-109.13 433.77 l
-111.18 432.28 l
-113.24 430.68 l
-115.30 428.95 l
-117.36 427.10 l
-119.42 425.10 l
-121.48 422.98 l
-123.54 420.71 l
-125.60 418.29 l
-127.66 415.74 l
-129.72 413.04 l
-131.77 410.20 l
-133.83 407.23 l
-135.89 404.12 l
-137.95 400.90 l
-140.01 397.57 l
-142.07 394.14 l
-144.13 390.63 l
-146.19 387.05 l
-148.25 383.43 l
-150.31 379.78 l
-152.36 376.12 l
-154.42 372.47 l
-156.48 368.86 l
-158.54 365.30 l
-160.60 361.82 l
-162.66 358.43 l
-164.72 355.15 l
-166.78 352.00 l
-168.84 348.99 l
-170.90 346.13 l
-172.96 343.42 l
-175.01 340.89 l
-177.07 338.52 l
-179.13 336.33 l
-181.19 334.30 l
-183.25 332.45 l
-185.31 330.76 l
-187.37 329.22 l
-189.43 327.84 l
-191.49 326.60 l
-193.55 325.49 l
-195.60 324.50 l
-197.66 323.63 l
-199.72 322.87 l
-201.78 322.20 l
-203.84 321.62 l
-205.90 321.11 l
-207.96 320.67 l
-210.02 320.29 l
-212.08 319.97 l
-214.14 319.69 l
-216.19 319.46 l
-218.25 319.26 l
-220.31 319.09 l
-S
-Q q
-0.000 0.000 0.000 RG
-0.75 w
-[] 0 d
-1 J
-1 j
-10.00 M
-55.59 312.96 m 220.31 312.96 l S
-55.59 312.96 m 55.59 306.98 l S
-96.77 312.96 m 96.77 306.98 l S
-137.95 312.96 m 137.95 306.98 l S
-179.13 312.96 m 179.13 306.98 l S
-220.31 312.96 m 220.31 306.98 l S
-BT
-0.000 0.000 0.000 rg
-/F2 1 Tf 10.00 0.00 -0.00 10.00 49.89 291.44 Tm (-4) Tj
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 91.07 291.44 Tm (-2) Tj
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 135.17 291.44 Tm (0) Tj
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 176.35 291.44 Tm (2) Tj
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 217.53 291.44 Tm (4) Tj
-ET
-49.00 318.22 m 49.00 449.74 l S
-49.00 318.22 m 43.03 318.22 l S
-49.00 344.52 m 43.03 344.52 l S
-49.00 370.82 m 43.03 370.82 l S
-49.00 397.13 m 43.03 397.13 l S
-49.00 423.43 m 43.03 423.43 l S
-49.00 449.74 m 43.03 449.74 l S
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 34.66 311.27 Tm (0.0) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 34.66 337.57 Tm (0.2) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 34.66 363.87 Tm (0.4) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 34.66 390.18 Tm (0.6) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 34.66 416.48 Tm (0.8) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 34.66 442.79 Tm (1.0) Tj
-ET
-49.00 312.96 m
-226.90 312.96 l
-226.90 455.00 l
-49.00 455.00 l
-49.00 312.96 l
-S
-Q q 0.00 252.00 252.00 252.00 re W n
-BT
-0.000 0.000 0.000 rg
-/F3 1 Tf 12.00 0.00 -0.00 12.00 83.06 475.19 Tm [(ICC plot f) 20 (or item I1)] TJ
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 99.05 267.54 Tm (Latent Dimension) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 10.76 341.31 Tm [(Probability to Solv) 25 (e)] TJ
-ET
-Q q 49.00 312.96 177.90 142.04 re W n
-1.000 0.000 0.000 RG
-0.75 w
-[] 0 d
-1 J
-1 j
-10.00 M
-55.59 319.53 m
-57.65 319.67 l
-59.71 319.82 l
-61.77 319.98 l
-63.83 320.16 l
-65.89 320.37 l
-67.95 320.59 l
-70.00 320.83 l
-72.06 321.10 l
-74.12 321.40 l
-76.18 321.72 l
-78.24 322.08 l
-80.30 322.48 l
-82.36 322.91 l
-84.42 323.38 l
-86.48 323.90 l
-88.54 324.47 l
-90.59 325.09 l
-92.65 325.77 l
-94.71 326.52 l
-96.77 327.33 l
-98.83 328.21 l
-100.89 329.17 l
-102.95 330.22 l
-105.01 331.35 l
-107.07 332.58 l
-109.13 333.90 l
-111.18 335.33 l
-113.24 336.86 l
-115.30 338.51 l
-117.36 340.26 l
-119.42 342.14 l
-121.48 344.13 l
-123.54 346.23 l
-125.60 348.45 l
-127.66 350.77 l
-129.72 353.20 l
-131.77 355.72 l
-133.83 358.33 l
-135.89 361.00 l
-137.95 363.73 l
-140.01 366.49 l
-142.07 369.26 l
-144.13 372.02 l
-146.19 374.74 l
-148.25 377.40 l
-150.31 379.96 l
-152.36 382.40 l
-154.42 384.68 l
-156.48 386.78 l
-158.54 388.67 l
-160.60 390.32 l
-162.66 391.71 l
-164.72 392.82 l
-166.78 393.63 l
-168.84 394.13 l
-170.90 394.31 l
-172.96 394.17 l
-175.01 393.72 l
-177.07 392.95 l
-179.13 391.88 l
-181.19 390.53 l
-183.25 388.92 l
-185.31 387.06 l
-187.37 384.99 l
-189.43 382.73 l
-191.49 380.32 l
-193.55 377.77 l
-195.60 375.13 l
-197.66 372.41 l
-199.72 369.65 l
-201.78 366.88 l
-203.84 364.12 l
-205.90 361.39 l
-207.96 358.71 l
-210.02 356.09 l
-212.08 353.56 l
-214.14 351.11 l
-216.19 348.77 l
-218.25 346.54 l
-220.31 344.42 l
-S
-0.000 0.804 0.000 RG
-55.59 318.22 m
-57.65 318.22 l
-59.71 318.22 l
-61.77 318.22 l
-63.83 318.22 l
-65.89 318.22 l
-67.95 318.22 l
-70.00 318.22 l
-72.06 318.22 l
-74.12 318.23 l
-76.18 318.23 l
-78.24 318.23 l
-80.30 318.23 l
-82.36 318.24 l
-84.42 318.24 l
-86.48 318.25 l
-88.54 318.26 l
-90.59 318.27 l
-92.65 318.28 l
-94.71 318.29 l
-96.77 318.31 l
-98.83 318.33 l
-100.89 318.35 l
-102.95 318.38 l
-105.01 318.41 l
-107.07 318.45 l
-109.13 318.50 l
-111.18 318.56 l
-113.24 318.63 l
-115.30 318.71 l
-117.36 318.81 l
-119.42 318.93 l
-121.48 319.07 l
-123.54 319.23 l
-125.60 319.43 l
-127.66 319.66 l
-129.72 319.93 l
-131.77 320.24 l
-133.83 320.61 l
-135.89 321.04 l
-137.95 321.54 l
-140.01 322.11 l
-142.07 322.77 l
-144.13 323.52 l
-146.19 324.37 l
-148.25 325.34 l
-150.31 326.43 l
-152.36 327.65 l
-154.42 329.02 l
-156.48 330.53 l
-158.54 332.20 l
-160.60 334.03 l
-162.66 336.03 l
-164.72 338.20 l
-166.78 340.54 l
-168.84 343.05 l
-170.90 345.73 l
-172.96 348.57 l
-175.01 351.56 l
-177.07 354.69 l
-179.13 357.96 l
-181.19 361.33 l
-183.25 364.80 l
-185.31 368.35 l
-187.37 371.95 l
-189.43 375.60 l
-191.49 379.26 l
-193.55 382.91 l
-195.60 386.54 l
-197.66 390.12 l
-199.72 393.64 l
-201.78 397.09 l
-203.84 400.43 l
-205.90 403.67 l
-207.96 406.79 l
-210.02 409.78 l
-212.08 412.64 l
-214.14 415.36 l
-216.19 417.94 l
-218.25 420.37 l
-220.31 422.66 l
-S
-Q q 301.00 312.96 177.90 142.04 re W n
-Q q 301.00 312.96 177.90 142.04 re W n
-0.000 0.000 0.000 RG
-0.75 w
-[] 0 d
-1 J
-1 j
-10.00 M
-307.59 448.77 m
-309.65 448.67 l
-311.71 448.56 l
-313.77 448.44 l
-315.83 448.31 l
-317.89 448.16 l
-319.95 447.99 l
-322.00 447.81 l
-324.06 447.61 l
-326.12 447.39 l
-328.18 447.15 l
-330.24 446.88 l
-332.30 446.59 l
-334.36 446.27 l
-336.42 445.91 l
-338.48 445.52 l
-340.54 445.09 l
-342.59 444.62 l
-344.65 444.10 l
-346.71 443.53 l
-348.77 442.91 l
-350.83 442.23 l
-352.89 441.48 l
-354.95 440.67 l
-357.01 439.78 l
-359.07 438.80 l
-361.13 437.74 l
-363.18 436.59 l
-365.24 435.34 l
-367.30 433.99 l
-369.36 432.52 l
-371.42 430.94 l
-373.48 429.23 l
-375.54 427.39 l
-377.60 425.42 l
-379.66 423.31 l
-381.72 421.06 l
-383.77 418.67 l
-385.83 416.14 l
-387.89 413.46 l
-389.95 410.64 l
-392.01 407.69 l
-394.07 404.61 l
-396.13 401.40 l
-398.19 398.09 l
-400.25 394.67 l
-402.31 391.17 l
-404.36 387.61 l
-406.42 383.99 l
-408.48 380.34 l
-410.54 376.68 l
-412.60 373.03 l
-414.66 369.41 l
-416.72 365.84 l
-418.78 362.35 l
-420.84 358.94 l
-422.90 355.65 l
-424.96 352.48 l
-427.01 349.44 l
-429.07 346.56 l
-431.13 343.83 l
-433.19 341.27 l
-435.25 338.88 l
-437.31 336.65 l
-439.37 334.60 l
-441.43 332.72 l
-443.49 331.01 l
-445.55 329.45 l
-447.60 328.04 l
-449.66 326.78 l
-451.72 325.65 l
-453.78 324.65 l
-455.84 323.76 l
-457.90 322.98 l
-459.96 322.30 l
-462.02 321.70 l
-464.08 321.18 l
-466.14 320.73 l
-468.19 320.35 l
-470.25 320.02 l
-472.31 319.73 l
-S
-Q q
-0.000 0.000 0.000 RG
-0.75 w
-[] 0 d
-1 J
-1 j
-10.00 M
-307.59 312.96 m 472.31 312.96 l S
-307.59 312.96 m 307.59 306.98 l S
-348.77 312.96 m 348.77 306.98 l S
-389.95 312.96 m 389.95 306.98 l S
-431.13 312.96 m 431.13 306.98 l S
-472.31 312.96 m 472.31 306.98 l S
-BT
-0.000 0.000 0.000 rg
-/F2 1 Tf 10.00 0.00 -0.00 10.00 301.89 291.44 Tm (-4) Tj
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 343.07 291.44 Tm (-2) Tj
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 387.17 291.44 Tm (0) Tj
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 428.35 291.44 Tm (2) Tj
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 469.53 291.44 Tm (4) Tj
-ET
-301.00 318.22 m 301.00 449.74 l S
-301.00 318.22 m 295.03 318.22 l S
-301.00 344.52 m 295.03 344.52 l S
-301.00 370.82 m 295.03 370.82 l S
-301.00 397.13 m 295.03 397.13 l S
-301.00 423.43 m 295.03 423.43 l S
-301.00 449.74 m 295.03 449.74 l S
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 286.66 311.27 Tm (0.0) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 286.66 337.57 Tm (0.2) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 286.66 363.87 Tm (0.4) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 286.66 390.18 Tm (0.6) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 286.66 416.48 Tm (0.8) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 286.66 442.79 Tm (1.0) Tj
-ET
-301.00 312.96 m
-478.90 312.96 l
-478.90 455.00 l
-301.00 455.00 l
-301.00 312.96 l
-S
-Q q 252.00 252.00 252.00 252.00 re W n
-BT
-0.000 0.000 0.000 rg
-/F3 1 Tf 12.00 0.00 -0.00 12.00 335.06 475.19 Tm [(ICC plot f) 20 (or item I2)] TJ
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 351.05 267.54 Tm (Latent Dimension) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 262.76 341.31 Tm [(Probability to Solv) 25 (e)] TJ
-ET
-Q q 301.00 312.96 177.90 142.04 re W n
-1.000 0.000 0.000 RG
-0.75 w
-[] 0 d
-1 J
-1 j
-10.00 M
-307.59 319.18 m
-309.65 319.28 l
-311.71 319.39 l
-313.77 319.51 l
-315.83 319.64 l
-317.89 319.79 l
-319.95 319.96 l
-322.00 320.14 l
-324.06 320.33 l
-326.12 320.55 l
-328.18 320.79 l
-330.24 321.06 l
-332.30 321.35 l
-334.36 321.67 l
-336.42 322.02 l
-338.48 322.41 l
-340.54 322.84 l
-342.59 323.30 l
-344.65 323.82 l
-346.71 324.38 l
-348.77 324.99 l
-350.83 325.66 l
-352.89 326.40 l
-354.95 327.20 l
-357.01 328.07 l
-359.07 329.02 l
-361.13 330.05 l
-363.18 331.17 l
-365.24 332.38 l
-367.30 333.69 l
-369.36 335.10 l
-371.42 336.62 l
-373.48 338.25 l
-375.54 339.99 l
-377.60 341.84 l
-379.66 343.81 l
-381.72 345.90 l
-383.77 348.10 l
-385.83 350.41 l
-387.89 352.82 l
-389.95 355.33 l
-392.01 357.92 l
-394.07 360.59 l
-396.13 363.31 l
-398.19 366.06 l
-400.25 368.83 l
-402.31 371.60 l
-404.36 374.33 l
-406.42 377.00 l
-408.48 379.57 l
-410.54 382.03 l
-412.60 384.34 l
-414.66 386.47 l
-416.72 388.39 l
-418.78 390.08 l
-420.84 391.51 l
-422.90 392.66 l
-424.96 393.52 l
-427.01 394.07 l
-429.07 394.30 l
-431.13 394.21 l
-433.19 393.81 l
-435.25 393.09 l
-437.31 392.07 l
-439.37 390.76 l
-441.43 389.18 l
-443.49 387.36 l
-445.55 385.32 l
-447.60 383.09 l
-449.66 380.70 l
-451.72 378.17 l
-453.78 375.54 l
-455.84 372.83 l
-457.90 370.08 l
-459.96 367.31 l
-462.02 364.54 l
-464.08 361.81 l
-466.14 359.12 l
-468.19 356.49 l
-470.25 353.94 l
-472.31 351.48 l
-S
-0.000 0.804 0.000 RG
-307.59 318.22 m
-309.65 318.22 l
-311.71 318.22 l
-313.77 318.22 l
-315.83 318.22 l
-317.89 318.22 l
-319.95 318.22 l
-322.00 318.22 l
-324.06 318.22 l
-326.12 318.22 l
-328.18 318.22 l
-330.24 318.22 l
-332.30 318.23 l
-334.36 318.23 l
-336.42 318.23 l
-338.48 318.23 l
-340.54 318.24 l
-342.59 318.24 l
-344.65 318.25 l
-346.71 318.26 l
-348.77 318.26 l
-350.83 318.28 l
-352.89 318.29 l
-354.95 318.30 l
-357.01 318.32 l
-359.07 318.34 l
-361.13 318.37 l
-363.18 318.40 l
-365.24 318.44 l
-367.30 318.49 l
-369.36 318.55 l
-371.42 318.61 l
-373.48 318.70 l
-375.54 318.79 l
-377.60 318.91 l
-379.66 319.04 l
-381.72 319.20 l
-383.77 319.40 l
-385.83 319.62 l
-387.89 319.88 l
-389.95 320.19 l
-392.01 320.55 l
-394.07 320.97 l
-396.13 321.46 l
-398.19 322.02 l
-400.25 322.66 l
-402.31 323.40 l
-404.36 324.23 l
-406.42 325.18 l
-408.48 326.25 l
-410.54 327.45 l
-412.60 328.80 l
-414.66 330.29 l
-416.72 331.93 l
-418.78 333.74 l
-420.84 335.71 l
-422.90 337.85 l
-424.96 340.17 l
-427.01 342.66 l
-429.07 345.31 l
-431.13 348.12 l
-433.19 351.09 l
-435.25 354.20 l
-437.31 357.45 l
-439.37 360.80 l
-441.43 364.26 l
-443.49 367.80 l
-445.55 371.40 l
-447.60 375.04 l
-449.66 378.69 l
-451.72 382.35 l
-453.78 385.98 l
-455.84 389.58 l
-457.90 393.11 l
-459.96 396.56 l
-462.02 399.92 l
-464.08 403.18 l
-466.14 406.32 l
-468.19 409.33 l
-470.25 412.21 l
-472.31 414.95 l
-S
-Q q 49.00 60.96 177.90 142.04 re W n
-Q q 49.00 60.96 177.90 142.04 re W n
-0.000 0.000 0.000 RG
-0.75 w
-[] 0 d
-1 J
-1 j
-10.00 M
-55.59 191.42 m
-57.65 190.78 l
-59.71 190.09 l
-61.77 189.33 l
-63.83 188.50 l
-65.89 187.59 l
-67.95 186.61 l
-70.00 185.53 l
-72.06 184.36 l
-74.12 183.09 l
-76.18 181.71 l
-78.24 180.22 l
-80.30 178.62 l
-82.36 176.88 l
-84.42 175.02 l
-86.48 173.03 l
-88.54 170.89 l
-90.59 168.62 l
-92.65 166.20 l
-94.71 163.64 l
-96.77 160.93 l
-98.83 158.09 l
-100.89 155.11 l
-102.95 152.00 l
-105.01 148.78 l
-107.07 145.44 l
-109.13 142.01 l
-111.18 138.49 l
-113.24 134.92 l
-115.30 131.29 l
-117.36 127.64 l
-119.42 123.98 l
-121.48 120.33 l
-123.54 116.72 l
-125.60 113.17 l
-127.66 109.69 l
-129.72 106.30 l
-131.77 103.03 l
-133.83 99.88 l
-135.89 96.88 l
-137.95 94.02 l
-140.01 91.32 l
-142.07 88.80 l
-144.13 86.44 l
-146.19 84.25 l
-148.25 82.23 l
-150.31 80.38 l
-152.36 78.69 l
-154.42 77.17 l
-156.48 75.79 l
-158.54 74.55 l
-160.60 73.45 l
-162.66 72.47 l
-164.72 71.60 l
-166.78 70.84 l
-168.84 70.18 l
-170.90 69.59 l
-172.96 69.09 l
-175.01 68.66 l
-177.07 68.28 l
-179.13 67.96 l
-181.19 67.68 l
-183.25 67.45 l
-185.31 67.25 l
-187.37 67.08 l
-189.43 66.94 l
-191.49 66.82 l
-193.55 66.72 l
-195.60 66.63 l
-197.66 66.56 l
-199.72 66.50 l
-201.78 66.45 l
-203.84 66.41 l
-205.90 66.38 l
-207.96 66.35 l
-210.02 66.33 l
-212.08 66.31 l
-214.14 66.29 l
-216.19 66.28 l
-218.25 66.27 l
-220.31 66.26 l
-S
-Q q
-0.000 0.000 0.000 RG
-0.75 w
-[] 0 d
-1 J
-1 j
-10.00 M
-55.59 60.96 m 220.31 60.96 l S
-55.59 60.96 m 55.59 54.98 l S
-96.77 60.96 m 96.77 54.98 l S
-137.95 60.96 m 137.95 54.98 l S
-179.13 60.96 m 179.13 54.98 l S
-220.31 60.96 m 220.31 54.98 l S
-BT
-0.000 0.000 0.000 rg
-/F2 1 Tf 10.00 0.00 -0.00 10.00 49.89 39.44 Tm (-4) Tj
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 91.07 39.44 Tm (-2) Tj
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 135.17 39.44 Tm (0) Tj
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 176.35 39.44 Tm (2) Tj
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 217.53 39.44 Tm (4) Tj
-ET
-49.00 66.22 m 49.00 197.74 l S
-49.00 66.22 m 43.03 66.22 l S
-49.00 92.52 m 43.03 92.52 l S
-49.00 118.82 m 43.03 118.82 l S
-49.00 145.13 m 43.03 145.13 l S
-49.00 171.43 m 43.03 171.43 l S
-49.00 197.74 m 43.03 197.74 l S
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 34.66 59.27 Tm (0.0) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 34.66 85.57 Tm (0.2) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 34.66 111.87 Tm (0.4) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 34.66 138.18 Tm (0.6) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 34.66 164.48 Tm (0.8) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 34.66 190.79 Tm (1.0) Tj
-ET
-49.00 60.96 m
-226.90 60.96 l
-226.90 203.00 l
-49.00 203.00 l
-49.00 60.96 l
-S
-Q q 0.00 0.00 252.00 252.00 re W n
-BT
-0.000 0.000 0.000 rg
-/F3 1 Tf 12.00 0.00 -0.00 12.00 83.06 223.19 Tm [(ICC plot f) 20 (or item I3)] TJ
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 99.05 15.54 Tm (Latent Dimension) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 10.76 89.31 Tm [(Probability to Solv) 25 (e)] TJ
-ET
-Q q 49.00 60.96 177.90 142.04 re W n
-1.000 0.000 0.000 RG
-0.75 w
-[] 0 d
-1 J
-1 j
-10.00 M
-55.59 72.49 m
-57.65 73.12 l
-59.71 73.80 l
-61.77 74.55 l
-63.83 75.36 l
-65.89 76.25 l
-67.95 77.21 l
-70.00 78.26 l
-72.06 79.40 l
-74.12 80.63 l
-76.18 81.95 l
-78.24 83.38 l
-80.30 84.92 l
-82.36 86.57 l
-84.42 88.33 l
-86.48 90.21 l
-88.54 92.20 l
-90.59 94.31 l
-92.65 96.53 l
-94.71 98.86 l
-96.77 101.30 l
-98.83 103.82 l
-100.89 106.43 l
-102.95 109.11 l
-105.01 111.83 l
-107.07 114.59 l
-109.13 117.37 l
-111.18 120.13 l
-113.24 122.85 l
-115.30 125.50 l
-117.36 128.06 l
-119.42 130.49 l
-121.48 132.76 l
-123.54 134.86 l
-125.60 136.74 l
-127.66 138.38 l
-129.72 139.76 l
-131.77 140.85 l
-133.83 141.65 l
-135.89 142.14 l
-137.95 142.31 l
-140.01 142.16 l
-142.07 141.69 l
-144.13 140.92 l
-146.19 139.84 l
-148.25 138.48 l
-150.31 136.85 l
-152.36 134.99 l
-154.42 132.91 l
-156.48 130.64 l
-158.54 128.22 l
-160.60 125.67 l
-162.66 123.02 l
-164.72 120.31 l
-166.78 117.55 l
-168.84 114.78 l
-170.90 112.02 l
-172.96 109.29 l
-175.01 106.61 l
-177.07 103.99 l
-179.13 101.46 l
-181.19 99.02 l
-183.25 96.69 l
-185.31 94.46 l
-187.37 92.34 l
-189.43 90.34 l
-191.49 88.45 l
-193.55 86.68 l
-195.60 85.03 l
-197.66 83.48 l
-199.72 82.04 l
-201.78 80.71 l
-203.84 79.47 l
-205.90 78.33 l
-207.96 77.28 l
-210.02 76.31 l
-212.08 75.42 l
-214.14 74.60 l
-216.19 73.85 l
-218.25 73.16 l
-220.31 72.53 l
-S
-0.000 0.804 0.000 RG
-55.59 66.26 m
-57.65 66.27 l
-59.71 66.28 l
-61.77 66.29 l
-63.83 66.31 l
-65.89 66.33 l
-67.95 66.35 l
-70.00 66.38 l
-72.06 66.41 l
-74.12 66.45 l
-76.18 66.50 l
-78.24 66.56 l
-80.30 66.63 l
-82.36 66.71 l
-84.42 66.81 l
-86.48 66.93 l
-88.54 67.07 l
-90.59 67.24 l
-92.65 67.44 l
-94.71 67.67 l
-96.77 67.94 l
-98.83 68.26 l
-100.89 68.63 l
-102.95 69.06 l
-105.01 69.56 l
-107.07 70.13 l
-109.13 70.79 l
-111.18 71.55 l
-113.24 72.41 l
-115.30 73.38 l
-117.36 74.47 l
-119.42 75.70 l
-121.48 77.07 l
-123.54 78.59 l
-125.60 80.26 l
-127.66 82.10 l
-129.72 84.11 l
-131.77 86.29 l
-133.83 88.63 l
-135.89 91.15 l
-137.95 93.84 l
-140.01 96.68 l
-142.07 99.68 l
-144.13 102.82 l
-146.19 106.08 l
-148.25 109.46 l
-150.31 112.93 l
-152.36 116.49 l
-154.42 120.09 l
-156.48 123.74 l
-158.54 127.40 l
-160.60 131.05 l
-162.66 134.68 l
-164.72 138.26 l
-166.78 141.78 l
-168.84 145.22 l
-170.90 148.56 l
-172.96 151.79 l
-175.01 154.91 l
-177.07 157.89 l
-179.13 160.75 l
-181.19 163.46 l
-183.25 166.03 l
-185.31 168.46 l
-187.37 170.75 l
-189.43 172.89 l
-191.49 174.89 l
-193.55 176.77 l
-195.60 178.51 l
-197.66 180.12 l
-199.72 181.62 l
-201.78 183.00 l
-203.84 184.28 l
-205.90 185.46 l
-207.96 186.54 l
-210.02 187.53 l
-212.08 188.44 l
-214.14 189.28 l
-216.19 190.04 l
-218.25 190.74 l
-220.31 191.38 l
-S
-Q q 301.00 60.96 177.90 142.04 re W n
-Q q 301.00 60.96 177.90 142.04 re W n
-0.000 0.000 0.000 RG
-0.75 w
-[] 0 d
-1 J
-1 j
-10.00 M
-307.59 193.86 m
-309.65 193.46 l
-311.71 193.02 l
-313.77 192.54 l
-315.83 192.02 l
-317.89 191.44 l
-319.95 190.81 l
-322.00 190.12 l
-324.06 189.36 l
-326.12 188.54 l
-328.18 187.63 l
-330.24 186.65 l
-332.30 185.58 l
-334.36 184.41 l
-336.42 183.15 l
-338.48 181.77 l
-340.54 180.29 l
-342.59 178.69 l
-344.65 176.96 l
-346.71 175.10 l
-348.77 173.11 l
-350.83 170.98 l
-352.89 168.71 l
-354.95 166.30 l
-357.01 163.74 l
-359.07 161.05 l
-361.13 158.21 l
-363.18 155.23 l
-365.24 152.13 l
-367.30 148.91 l
-369.36 145.58 l
-371.42 142.15 l
-373.48 138.64 l
-375.54 135.06 l
-377.60 131.44 l
-379.66 127.79 l
-381.72 124.13 l
-383.77 120.48 l
-385.83 116.87 l
-387.89 113.31 l
-389.95 109.83 l
-392.01 106.44 l
-394.07 103.16 l
-396.13 100.01 l
-398.19 97.00 l
-400.25 94.13 l
-402.31 91.43 l
-404.36 88.90 l
-406.42 86.53 l
-408.48 84.33 l
-410.54 82.31 l
-412.60 80.45 l
-414.66 78.76 l
-416.72 77.23 l
-418.78 75.84 l
-420.84 74.60 l
-422.90 73.49 l
-424.96 72.51 l
-427.01 71.64 l
-429.07 70.87 l
-431.13 70.20 l
-433.19 69.62 l
-435.25 69.11 l
-437.31 68.67 l
-439.37 68.29 l
-441.43 67.97 l
-443.49 67.69 l
-445.55 67.46 l
-447.60 67.26 l
-449.66 67.09 l
-451.72 66.94 l
-453.78 66.82 l
-455.84 66.72 l
-457.90 66.64 l
-459.96 66.57 l
-462.02 66.51 l
-464.08 66.46 l
-466.14 66.42 l
-468.19 66.38 l
-470.25 66.35 l
-472.31 66.33 l
-S
-Q q
-0.000 0.000 0.000 RG
-0.75 w
-[] 0 d
-1 J
-1 j
-10.00 M
-307.59 60.96 m 472.31 60.96 l S
-307.59 60.96 m 307.59 54.98 l S
-348.77 60.96 m 348.77 54.98 l S
-389.95 60.96 m 389.95 54.98 l S
-431.13 60.96 m 431.13 54.98 l S
-472.31 60.96 m 472.31 54.98 l S
-BT
-0.000 0.000 0.000 rg
-/F2 1 Tf 10.00 0.00 -0.00 10.00 301.89 39.44 Tm (-4) Tj
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 343.07 39.44 Tm (-2) Tj
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 387.17 39.44 Tm (0) Tj
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 428.35 39.44 Tm (2) Tj
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 469.53 39.44 Tm (4) Tj
-ET
-301.00 66.22 m 301.00 197.74 l S
-301.00 66.22 m 295.03 66.22 l S
-301.00 92.52 m 295.03 92.52 l S
-301.00 118.82 m 295.03 118.82 l S
-301.00 145.13 m 295.03 145.13 l S
-301.00 171.43 m 295.03 171.43 l S
-301.00 197.74 m 295.03 197.74 l S
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 286.66 59.27 Tm (0.0) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 286.66 85.57 Tm (0.2) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 286.66 111.87 Tm (0.4) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 286.66 138.18 Tm (0.6) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 286.66 164.48 Tm (0.8) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 286.66 190.79 Tm (1.0) Tj
-ET
-301.00 60.96 m
-478.90 60.96 l
-478.90 203.00 l
-301.00 203.00 l
-301.00 60.96 l
-S
-Q q 252.00 0.00 252.00 252.00 re W n
-BT
-0.000 0.000 0.000 rg
-/F3 1 Tf 12.00 0.00 -0.00 12.00 335.06 223.19 Tm [(ICC plot f) 20 (or item I4)] TJ
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 351.05 15.54 Tm (Latent Dimension) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 262.76 89.31 Tm [(Probability to Solv) 25 (e)] TJ
-ET
-Q q 301.00 60.96 177.90 142.04 re W n
-1.000 0.000 0.000 RG
-0.75 w
-[] 0 d
-1 J
-1 j
-10.00 M
-307.59 70.08 m
-309.65 70.47 l
-311.71 70.91 l
-313.77 71.38 l
-315.83 71.90 l
-317.89 72.47 l
-319.95 73.09 l
-322.00 73.77 l
-324.06 74.51 l
-326.12 75.33 l
-328.18 76.21 l
-330.24 77.17 l
-332.30 78.21 l
-334.36 79.35 l
-336.42 80.57 l
-338.48 81.90 l
-340.54 83.32 l
-342.59 84.86 l
-344.65 86.50 l
-346.71 88.26 l
-348.77 90.13 l
-350.83 92.12 l
-352.89 94.22 l
-354.95 96.44 l
-357.01 98.77 l
-359.07 101.19 l
-361.13 103.72 l
-363.18 106.32 l
-365.24 108.99 l
-367.30 111.72 l
-369.36 114.48 l
-371.42 117.25 l
-373.48 120.01 l
-375.54 122.73 l
-377.60 125.39 l
-379.66 127.95 l
-381.72 130.39 l
-383.77 132.67 l
-385.83 134.77 l
-387.89 136.66 l
-389.95 138.31 l
-392.01 139.70 l
-394.07 140.81 l
-396.13 141.62 l
-398.19 142.13 l
-400.25 142.31 l
-402.31 142.17 l
-404.36 141.72 l
-406.42 140.95 l
-408.48 139.89 l
-410.54 138.54 l
-412.60 136.92 l
-414.66 135.07 l
-416.72 133.00 l
-418.78 130.74 l
-420.84 128.32 l
-422.90 125.78 l
-424.96 123.13 l
-427.01 120.42 l
-429.07 117.66 l
-431.13 114.89 l
-433.19 112.13 l
-435.25 109.40 l
-437.31 106.71 l
-439.37 104.10 l
-441.43 101.56 l
-443.49 99.12 l
-445.55 96.78 l
-447.60 94.55 l
-449.66 92.43 l
-451.72 90.42 l
-453.78 88.53 l
-455.84 86.75 l
-457.90 85.09 l
-459.96 83.54 l
-462.02 82.10 l
-464.08 80.76 l
-466.14 79.52 l
-468.19 78.38 l
-470.25 77.32 l
-472.31 76.35 l
-S
-0.000 0.804 0.000 RG
-307.59 66.23 m
-309.65 66.23 l
-311.71 66.24 l
-313.77 66.24 l
-315.83 66.25 l
-317.89 66.26 l
-319.95 66.27 l
-322.00 66.28 l
-324.06 66.29 l
-326.12 66.31 l
-328.18 66.33 l
-330.24 66.35 l
-332.30 66.38 l
-334.36 66.41 l
-336.42 66.45 l
-338.48 66.50 l
-340.54 66.56 l
-342.59 66.63 l
-344.65 66.71 l
-346.71 66.81 l
-348.77 66.93 l
-350.83 67.07 l
-352.89 67.23 l
-354.95 67.43 l
-357.01 67.66 l
-359.07 67.93 l
-361.13 68.24 l
-363.18 68.61 l
-365.24 69.04 l
-367.30 69.54 l
-369.36 70.11 l
-371.42 70.77 l
-373.48 71.52 l
-375.54 72.37 l
-377.60 73.34 l
-379.66 74.43 l
-381.72 75.65 l
-383.77 77.01 l
-385.83 78.52 l
-387.89 80.19 l
-389.95 82.02 l
-392.01 84.02 l
-394.07 86.19 l
-396.13 88.53 l
-398.19 91.05 l
-400.25 93.72 l
-402.31 96.56 l
-404.36 99.55 l
-406.42 102.68 l
-408.48 105.95 l
-410.54 109.32 l
-412.60 112.79 l
-414.66 116.34 l
-416.72 119.94 l
-418.78 123.59 l
-420.84 127.25 l
-422.90 130.90 l
-424.96 134.53 l
-427.01 138.11 l
-429.07 141.63 l
-431.13 145.08 l
-433.19 148.42 l
-435.25 151.66 l
-437.31 154.78 l
-439.37 157.77 l
-441.43 160.63 l
-443.49 163.35 l
-445.55 165.93 l
-447.60 168.36 l
-449.66 170.65 l
-451.72 172.80 l
-453.78 174.81 l
-455.84 176.69 l
-457.90 178.44 l
-459.96 180.06 l
-462.02 181.56 l
-464.08 182.95 l
-466.14 184.23 l
-468.19 185.41 l
-470.25 186.50 l
-472.31 187.49 l
-S
-Q
-endstream
-endobj
-10 0 obj
-22469
-endobj
-11 0 obj
-<<
-/Type /Page
-/Parent 3 0 R
-/Contents 12 0 R
-/Resources 4 0 R
->>
-endobj
-12 0 obj
-<<
-/Length 13 0 R
->>
-stream
-1 J 1 j q
-Q q 49.00 312.96 177.90 142.04 re W n
-Q q 49.00 312.96 177.90 142.04 re W n
-0.000 0.000 0.000 RG
-0.75 w
-[] 0 d
-1 J
-1 j
-10.00 M
-55.59 448.42 m
-57.65 448.28 l
-59.71 448.13 l
-61.77 447.97 l
-63.83 447.78 l
-65.89 447.58 l
-67.95 447.36 l
-70.00 447.11 l
-72.06 446.84 l
-74.12 446.54 l
-76.18 446.22 l
-78.24 445.85 l
-80.30 445.46 l
-82.36 445.02 l
-84.42 444.54 l
-86.48 444.02 l
-88.54 443.44 l
-90.59 442.81 l
-92.65 442.12 l
-94.71 441.36 l
-96.77 440.53 l
-98.83 439.63 l
-100.89 438.65 l
-102.95 437.57 l
-105.01 436.41 l
-107.07 435.14 l
-109.13 433.77 l
-111.18 432.28 l
-113.24 430.68 l
-115.30 428.95 l
-117.36 427.10 l
-119.42 425.10 l
-121.48 422.98 l
-123.54 420.71 l
-125.60 418.29 l
-127.66 415.74 l
-129.72 413.04 l
-131.77 410.20 l
-133.83 407.23 l
-135.89 404.12 l
-137.95 400.90 l
-140.01 397.57 l
-142.07 394.14 l
-144.13 390.63 l
-146.19 387.05 l
-148.25 383.43 l
-150.31 379.78 l
-152.36 376.12 l
-154.42 372.47 l
-156.48 368.86 l
-158.54 365.30 l
-160.60 361.82 l
-162.66 358.43 l
-164.72 355.15 l
-166.78 352.00 l
-168.84 348.99 l
-170.90 346.13 l
-172.96 343.42 l
-175.01 340.89 l
-177.07 338.52 l
-179.13 336.33 l
-181.19 334.30 l
-183.25 332.45 l
-185.31 330.76 l
-187.37 329.22 l
-189.43 327.84 l
-191.49 326.60 l
-193.55 325.49 l
-195.60 324.50 l
-197.66 323.63 l
-199.72 322.87 l
-201.78 322.20 l
-203.84 321.62 l
-205.90 321.11 l
-207.96 320.67 l
-210.02 320.29 l
-212.08 319.97 l
-214.14 319.69 l
-216.19 319.46 l
-218.25 319.26 l
-220.31 319.09 l
-S
-Q q
-0.000 0.000 0.000 RG
-0.75 w
-[] 0 d
-1 J
-1 j
-10.00 M
-55.59 312.96 m 220.31 312.96 l S
-55.59 312.96 m 55.59 306.98 l S
-96.77 312.96 m 96.77 306.98 l S
-137.95 312.96 m 137.95 306.98 l S
-179.13 312.96 m 179.13 306.98 l S
-220.31 312.96 m 220.31 306.98 l S
-BT
-0.000 0.000 0.000 rg
-/F2 1 Tf 10.00 0.00 -0.00 10.00 49.89 291.44 Tm (-4) Tj
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 91.07 291.44 Tm (-2) Tj
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 135.17 291.44 Tm (0) Tj
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 176.35 291.44 Tm (2) Tj
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 217.53 291.44 Tm (4) Tj
-ET
-49.00 318.22 m 49.00 449.74 l S
-49.00 318.22 m 43.03 318.22 l S
-49.00 344.52 m 43.03 344.52 l S
-49.00 370.82 m 43.03 370.82 l S
-49.00 397.13 m 43.03 397.13 l S
-49.00 423.43 m 43.03 423.43 l S
-49.00 449.74 m 43.03 449.74 l S
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 34.66 311.27 Tm (0.0) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 34.66 337.57 Tm (0.2) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 34.66 363.87 Tm (0.4) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 34.66 390.18 Tm (0.6) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 34.66 416.48 Tm (0.8) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 34.66 442.79 Tm (1.0) Tj
-ET
-49.00 312.96 m
-226.90 312.96 l
-226.90 455.00 l
-49.00 455.00 l
-49.00 312.96 l
-S
-Q q 0.00 252.00 252.00 252.00 re W n
-BT
-0.000 0.000 0.000 rg
-/F3 1 Tf 12.00 0.00 -0.00 12.00 83.06 475.19 Tm [(ICC plot f) 20 (or item I1)] TJ
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 99.05 267.54 Tm (Latent Dimension) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 10.76 341.31 Tm [(Probability to Solv) 25 (e)] TJ
-ET
-Q q 49.00 312.96 177.90 142.04 re W n
-1.000 0.000 0.000 RG
-0.75 w
-[] 0 d
-1 J
-1 j
-10.00 M
-55.59 319.53 m
-57.65 319.67 l
-59.71 319.82 l
-61.77 319.98 l
-63.83 320.16 l
-65.89 320.37 l
-67.95 320.59 l
-70.00 320.83 l
-72.06 321.10 l
-74.12 321.40 l
-76.18 321.72 l
-78.24 322.08 l
-80.30 322.48 l
-82.36 322.91 l
-84.42 323.38 l
-86.48 323.90 l
-88.54 324.47 l
-90.59 325.09 l
-92.65 325.77 l
-94.71 326.52 l
-96.77 327.33 l
-98.83 328.21 l
-100.89 329.17 l
-102.95 330.22 l
-105.01 331.35 l
-107.07 332.58 l
-109.13 333.90 l
-111.18 335.33 l
-113.24 336.86 l
-115.30 338.51 l
-117.36 340.26 l
-119.42 342.14 l
-121.48 344.13 l
-123.54 346.23 l
-125.60 348.45 l
-127.66 350.77 l
-129.72 353.20 l
-131.77 355.72 l
-133.83 358.33 l
-135.89 361.00 l
-137.95 363.73 l
-140.01 366.49 l
-142.07 369.26 l
-144.13 372.02 l
-146.19 374.74 l
-148.25 377.40 l
-150.31 379.96 l
-152.36 382.40 l
-154.42 384.68 l
-156.48 386.78 l
-158.54 388.67 l
-160.60 390.32 l
-162.66 391.71 l
-164.72 392.82 l
-166.78 393.63 l
-168.84 394.13 l
-170.90 394.31 l
-172.96 394.17 l
-175.01 393.72 l
-177.07 392.95 l
-179.13 391.88 l
-181.19 390.53 l
-183.25 388.92 l
-185.31 387.06 l
-187.37 384.99 l
-189.43 382.73 l
-191.49 380.32 l
-193.55 377.77 l
-195.60 375.13 l
-197.66 372.41 l
-199.72 369.65 l
-201.78 366.88 l
-203.84 364.12 l
-205.90 361.39 l
-207.96 358.71 l
-210.02 356.09 l
-212.08 353.56 l
-214.14 351.11 l
-216.19 348.77 l
-218.25 346.54 l
-220.31 344.42 l
-S
-0.000 0.804 0.000 RG
-55.59 318.22 m
-57.65 318.22 l
-59.71 318.22 l
-61.77 318.22 l
-63.83 318.22 l
-65.89 318.22 l
-67.95 318.22 l
-70.00 318.22 l
-72.06 318.22 l
-74.12 318.23 l
-76.18 318.23 l
-78.24 318.23 l
-80.30 318.23 l
-82.36 318.24 l
-84.42 318.24 l
-86.48 318.25 l
-88.54 318.26 l
-90.59 318.27 l
-92.65 318.28 l
-94.71 318.29 l
-96.77 318.31 l
-98.83 318.33 l
-100.89 318.35 l
-102.95 318.38 l
-105.01 318.41 l
-107.07 318.45 l
-109.13 318.50 l
-111.18 318.56 l
-113.24 318.63 l
-115.30 318.71 l
-117.36 318.81 l
-119.42 318.93 l
-121.48 319.07 l
-123.54 319.23 l
-125.60 319.43 l
-127.66 319.66 l
-129.72 319.93 l
-131.77 320.24 l
-133.83 320.61 l
-135.89 321.04 l
-137.95 321.54 l
-140.01 322.11 l
-142.07 322.77 l
-144.13 323.52 l
-146.19 324.37 l
-148.25 325.34 l
-150.31 326.43 l
-152.36 327.65 l
-154.42 329.02 l
-156.48 330.53 l
-158.54 332.20 l
-160.60 334.03 l
-162.66 336.03 l
-164.72 338.20 l
-166.78 340.54 l
-168.84 343.05 l
-170.90 345.73 l
-172.96 348.57 l
-175.01 351.56 l
-177.07 354.69 l
-179.13 357.96 l
-181.19 361.33 l
-183.25 364.80 l
-185.31 368.35 l
-187.37 371.95 l
-189.43 375.60 l
-191.49 379.26 l
-193.55 382.91 l
-195.60 386.54 l
-197.66 390.12 l
-199.72 393.64 l
-201.78 397.09 l
-203.84 400.43 l
-205.90 403.67 l
-207.96 406.79 l
-210.02 409.78 l
-212.08 412.64 l
-214.14 415.36 l
-216.19 417.94 l
-218.25 420.37 l
-220.31 422.66 l
-S
-Q q 301.00 312.96 177.90 142.04 re W n
-Q q 301.00 312.96 177.90 142.04 re W n
-0.000 0.000 0.000 RG
-0.75 w
-[] 0 d
-1 J
-1 j
-10.00 M
-307.59 448.77 m
-309.65 448.67 l
-311.71 448.56 l
-313.77 448.44 l
-315.83 448.31 l
-317.89 448.16 l
-319.95 447.99 l
-322.00 447.81 l
-324.06 447.61 l
-326.12 447.39 l
-328.18 447.15 l
-330.24 446.88 l
-332.30 446.59 l
-334.36 446.27 l
-336.42 445.91 l
-338.48 445.52 l
-340.54 445.09 l
-342.59 444.62 l
-344.65 444.10 l
-346.71 443.53 l
-348.77 442.91 l
-350.83 442.23 l
-352.89 441.48 l
-354.95 440.67 l
-357.01 439.78 l
-359.07 438.80 l
-361.13 437.74 l
-363.18 436.59 l
-365.24 435.34 l
-367.30 433.99 l
-369.36 432.52 l
-371.42 430.94 l
-373.48 429.23 l
-375.54 427.39 l
-377.60 425.42 l
-379.66 423.31 l
-381.72 421.06 l
-383.77 418.67 l
-385.83 416.14 l
-387.89 413.46 l
-389.95 410.64 l
-392.01 407.69 l
-394.07 404.61 l
-396.13 401.40 l
-398.19 398.09 l
-400.25 394.67 l
-402.31 391.17 l
-404.36 387.61 l
-406.42 383.99 l
-408.48 380.34 l
-410.54 376.68 l
-412.60 373.03 l
-414.66 369.41 l
-416.72 365.84 l
-418.78 362.35 l
-420.84 358.94 l
-422.90 355.65 l
-424.96 352.48 l
-427.01 349.44 l
-429.07 346.56 l
-431.13 343.83 l
-433.19 341.27 l
-435.25 338.88 l
-437.31 336.65 l
-439.37 334.60 l
-441.43 332.72 l
-443.49 331.01 l
-445.55 329.45 l
-447.60 328.04 l
-449.66 326.78 l
-451.72 325.65 l
-453.78 324.65 l
-455.84 323.76 l
-457.90 322.98 l
-459.96 322.30 l
-462.02 321.70 l
-464.08 321.18 l
-466.14 320.73 l
-468.19 320.35 l
-470.25 320.02 l
-472.31 319.73 l
-S
-Q q
-0.000 0.000 0.000 RG
-0.75 w
-[] 0 d
-1 J
-1 j
-10.00 M
-307.59 312.96 m 472.31 312.96 l S
-307.59 312.96 m 307.59 306.98 l S
-348.77 312.96 m 348.77 306.98 l S
-389.95 312.96 m 389.95 306.98 l S
-431.13 312.96 m 431.13 306.98 l S
-472.31 312.96 m 472.31 306.98 l S
-BT
-0.000 0.000 0.000 rg
-/F2 1 Tf 10.00 0.00 -0.00 10.00 301.89 291.44 Tm (-4) Tj
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 343.07 291.44 Tm (-2) Tj
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 387.17 291.44 Tm (0) Tj
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 428.35 291.44 Tm (2) Tj
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 469.53 291.44 Tm (4) Tj
-ET
-301.00 318.22 m 301.00 449.74 l S
-301.00 318.22 m 295.03 318.22 l S
-301.00 344.52 m 295.03 344.52 l S
-301.00 370.82 m 295.03 370.82 l S
-301.00 397.13 m 295.03 397.13 l S
-301.00 423.43 m 295.03 423.43 l S
-301.00 449.74 m 295.03 449.74 l S
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 286.66 311.27 Tm (0.0) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 286.66 337.57 Tm (0.2) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 286.66 363.87 Tm (0.4) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 286.66 390.18 Tm (0.6) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 286.66 416.48 Tm (0.8) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 286.66 442.79 Tm (1.0) Tj
-ET
-301.00 312.96 m
-478.90 312.96 l
-478.90 455.00 l
-301.00 455.00 l
-301.00 312.96 l
-S
-Q q 252.00 252.00 252.00 252.00 re W n
-BT
-0.000 0.000 0.000 rg
-/F3 1 Tf 12.00 0.00 -0.00 12.00 335.06 475.19 Tm [(ICC plot f) 20 (or item I2)] TJ
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 351.05 267.54 Tm (Latent Dimension) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 262.76 341.31 Tm [(Probability to Solv) 25 (e)] TJ
-ET
-Q q 301.00 312.96 177.90 142.04 re W n
-1.000 0.000 0.000 RG
-0.75 w
-[] 0 d
-1 J
-1 j
-10.00 M
-307.59 319.18 m
-309.65 319.28 l
-311.71 319.39 l
-313.77 319.51 l
-315.83 319.64 l
-317.89 319.79 l
-319.95 319.96 l
-322.00 320.14 l
-324.06 320.33 l
-326.12 320.55 l
-328.18 320.79 l
-330.24 321.06 l
-332.30 321.35 l
-334.36 321.67 l
-336.42 322.02 l
-338.48 322.41 l
-340.54 322.84 l
-342.59 323.30 l
-344.65 323.82 l
-346.71 324.38 l
-348.77 324.99 l
-350.83 325.66 l
-352.89 326.40 l
-354.95 327.20 l
-357.01 328.07 l
-359.07 329.02 l
-361.13 330.05 l
-363.18 331.17 l
-365.24 332.38 l
-367.30 333.69 l
-369.36 335.10 l
-371.42 336.62 l
-373.48 338.25 l
-375.54 339.99 l
-377.60 341.84 l
-379.66 343.81 l
-381.72 345.90 l
-383.77 348.10 l
-385.83 350.41 l
-387.89 352.82 l
-389.95 355.33 l
-392.01 357.92 l
-394.07 360.59 l
-396.13 363.31 l
-398.19 366.06 l
-400.25 368.83 l
-402.31 371.60 l
-404.36 374.33 l
-406.42 377.00 l
-408.48 379.57 l
-410.54 382.03 l
-412.60 384.34 l
-414.66 386.47 l
-416.72 388.39 l
-418.78 390.08 l
-420.84 391.51 l
-422.90 392.66 l
-424.96 393.52 l
-427.01 394.07 l
-429.07 394.30 l
-431.13 394.21 l
-433.19 393.81 l
-435.25 393.09 l
-437.31 392.07 l
-439.37 390.76 l
-441.43 389.18 l
-443.49 387.36 l
-445.55 385.32 l
-447.60 383.09 l
-449.66 380.70 l
-451.72 378.17 l
-453.78 375.54 l
-455.84 372.83 l
-457.90 370.08 l
-459.96 367.31 l
-462.02 364.54 l
-464.08 361.81 l
-466.14 359.12 l
-468.19 356.49 l
-470.25 353.94 l
-472.31 351.48 l
-S
-0.000 0.804 0.000 RG
-307.59 318.22 m
-309.65 318.22 l
-311.71 318.22 l
-313.77 318.22 l
-315.83 318.22 l
-317.89 318.22 l
-319.95 318.22 l
-322.00 318.22 l
-324.06 318.22 l
-326.12 318.22 l
-328.18 318.22 l
-330.24 318.22 l
-332.30 318.23 l
-334.36 318.23 l
-336.42 318.23 l
-338.48 318.23 l
-340.54 318.24 l
-342.59 318.24 l
-344.65 318.25 l
-346.71 318.26 l
-348.77 318.26 l
-350.83 318.28 l
-352.89 318.29 l
-354.95 318.30 l
-357.01 318.32 l
-359.07 318.34 l
-361.13 318.37 l
-363.18 318.40 l
-365.24 318.44 l
-367.30 318.49 l
-369.36 318.55 l
-371.42 318.61 l
-373.48 318.70 l
-375.54 318.79 l
-377.60 318.91 l
-379.66 319.04 l
-381.72 319.20 l
-383.77 319.40 l
-385.83 319.62 l
-387.89 319.88 l
-389.95 320.19 l
-392.01 320.55 l
-394.07 320.97 l
-396.13 321.46 l
-398.19 322.02 l
-400.25 322.66 l
-402.31 323.40 l
-404.36 324.23 l
-406.42 325.18 l
-408.48 326.25 l
-410.54 327.45 l
-412.60 328.80 l
-414.66 330.29 l
-416.72 331.93 l
-418.78 333.74 l
-420.84 335.71 l
-422.90 337.85 l
-424.96 340.17 l
-427.01 342.66 l
-429.07 345.31 l
-431.13 348.12 l
-433.19 351.09 l
-435.25 354.20 l
-437.31 357.45 l
-439.37 360.80 l
-441.43 364.26 l
-443.49 367.80 l
-445.55 371.40 l
-447.60 375.04 l
-449.66 378.69 l
-451.72 382.35 l
-453.78 385.98 l
-455.84 389.58 l
-457.90 393.11 l
-459.96 396.56 l
-462.02 399.92 l
-464.08 403.18 l
-466.14 406.32 l
-468.19 409.33 l
-470.25 412.21 l
-472.31 414.95 l
-S
-Q q 49.00 60.96 177.90 142.04 re W n
-Q q 49.00 60.96 177.90 142.04 re W n
-0.000 0.000 0.000 RG
-0.75 w
-[] 0 d
-1 J
-1 j
-10.00 M
-55.59 191.42 m
-57.65 190.78 l
-59.71 190.09 l
-61.77 189.33 l
-63.83 188.50 l
-65.89 187.59 l
-67.95 186.61 l
-70.00 185.53 l
-72.06 184.36 l
-74.12 183.09 l
-76.18 181.71 l
-78.24 180.22 l
-80.30 178.62 l
-82.36 176.88 l
-84.42 175.02 l
-86.48 173.03 l
-88.54 170.89 l
-90.59 168.62 l
-92.65 166.20 l
-94.71 163.64 l
-96.77 160.93 l
-98.83 158.09 l
-100.89 155.11 l
-102.95 152.00 l
-105.01 148.78 l
-107.07 145.44 l
-109.13 142.01 l
-111.18 138.49 l
-113.24 134.92 l
-115.30 131.29 l
-117.36 127.64 l
-119.42 123.98 l
-121.48 120.33 l
-123.54 116.72 l
-125.60 113.17 l
-127.66 109.69 l
-129.72 106.30 l
-131.77 103.03 l
-133.83 99.88 l
-135.89 96.88 l
-137.95 94.02 l
-140.01 91.32 l
-142.07 88.80 l
-144.13 86.44 l
-146.19 84.25 l
-148.25 82.23 l
-150.31 80.38 l
-152.36 78.69 l
-154.42 77.17 l
-156.48 75.79 l
-158.54 74.55 l
-160.60 73.45 l
-162.66 72.47 l
-164.72 71.60 l
-166.78 70.84 l
-168.84 70.18 l
-170.90 69.59 l
-172.96 69.09 l
-175.01 68.66 l
-177.07 68.28 l
-179.13 67.96 l
-181.19 67.68 l
-183.25 67.45 l
-185.31 67.25 l
-187.37 67.08 l
-189.43 66.94 l
-191.49 66.82 l
-193.55 66.72 l
-195.60 66.63 l
-197.66 66.56 l
-199.72 66.50 l
-201.78 66.45 l
-203.84 66.41 l
-205.90 66.38 l
-207.96 66.35 l
-210.02 66.33 l
-212.08 66.31 l
-214.14 66.29 l
-216.19 66.28 l
-218.25 66.27 l
-220.31 66.26 l
-S
-Q q
-0.000 0.000 0.000 RG
-0.75 w
-[] 0 d
-1 J
-1 j
-10.00 M
-55.59 60.96 m 220.31 60.96 l S
-55.59 60.96 m 55.59 54.98 l S
-96.77 60.96 m 96.77 54.98 l S
-137.95 60.96 m 137.95 54.98 l S
-179.13 60.96 m 179.13 54.98 l S
-220.31 60.96 m 220.31 54.98 l S
-BT
-0.000 0.000 0.000 rg
-/F2 1 Tf 10.00 0.00 -0.00 10.00 49.89 39.44 Tm (-4) Tj
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 91.07 39.44 Tm (-2) Tj
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 135.17 39.44 Tm (0) Tj
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 176.35 39.44 Tm (2) Tj
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 217.53 39.44 Tm (4) Tj
-ET
-49.00 66.22 m 49.00 197.74 l S
-49.00 66.22 m 43.03 66.22 l S
-49.00 92.52 m 43.03 92.52 l S
-49.00 118.82 m 43.03 118.82 l S
-49.00 145.13 m 43.03 145.13 l S
-49.00 171.43 m 43.03 171.43 l S
-49.00 197.74 m 43.03 197.74 l S
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 34.66 59.27 Tm (0.0) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 34.66 85.57 Tm (0.2) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 34.66 111.87 Tm (0.4) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 34.66 138.18 Tm (0.6) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 34.66 164.48 Tm (0.8) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 34.66 190.79 Tm (1.0) Tj
-ET
-49.00 60.96 m
-226.90 60.96 l
-226.90 203.00 l
-49.00 203.00 l
-49.00 60.96 l
-S
-Q q 0.00 0.00 252.00 252.00 re W n
-BT
-0.000 0.000 0.000 rg
-/F3 1 Tf 12.00 0.00 -0.00 12.00 83.06 223.19 Tm [(ICC plot f) 20 (or item I3)] TJ
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 99.05 15.54 Tm (Latent Dimension) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 10.76 89.31 Tm [(Probability to Solv) 25 (e)] TJ
-ET
-Q q 49.00 60.96 177.90 142.04 re W n
-1.000 0.000 0.000 RG
-0.75 w
-[] 0 d
-1 J
-1 j
-10.00 M
-55.59 72.49 m
-57.65 73.12 l
-59.71 73.80 l
-61.77 74.55 l
-63.83 75.36 l
-65.89 76.25 l
-67.95 77.21 l
-70.00 78.26 l
-72.06 79.40 l
-74.12 80.63 l
-76.18 81.95 l
-78.24 83.38 l
-80.30 84.92 l
-82.36 86.57 l
-84.42 88.33 l
-86.48 90.21 l
-88.54 92.20 l
-90.59 94.31 l
-92.65 96.53 l
-94.71 98.86 l
-96.77 101.30 l
-98.83 103.82 l
-100.89 106.43 l
-102.95 109.11 l
-105.01 111.83 l
-107.07 114.59 l
-109.13 117.37 l
-111.18 120.13 l
-113.24 122.85 l
-115.30 125.50 l
-117.36 128.06 l
-119.42 130.49 l
-121.48 132.76 l
-123.54 134.86 l
-125.60 136.74 l
-127.66 138.38 l
-129.72 139.76 l
-131.77 140.85 l
-133.83 141.65 l
-135.89 142.14 l
-137.95 142.31 l
-140.01 142.16 l
-142.07 141.69 l
-144.13 140.92 l
-146.19 139.84 l
-148.25 138.48 l
-150.31 136.85 l
-152.36 134.99 l
-154.42 132.91 l
-156.48 130.64 l
-158.54 128.22 l
-160.60 125.67 l
-162.66 123.02 l
-164.72 120.31 l
-166.78 117.55 l
-168.84 114.78 l
-170.90 112.02 l
-172.96 109.29 l
-175.01 106.61 l
-177.07 103.99 l
-179.13 101.46 l
-181.19 99.02 l
-183.25 96.69 l
-185.31 94.46 l
-187.37 92.34 l
-189.43 90.34 l
-191.49 88.45 l
-193.55 86.68 l
-195.60 85.03 l
-197.66 83.48 l
-199.72 82.04 l
-201.78 80.71 l
-203.84 79.47 l
-205.90 78.33 l
-207.96 77.28 l
-210.02 76.31 l
-212.08 75.42 l
-214.14 74.60 l
-216.19 73.85 l
-218.25 73.16 l
-220.31 72.53 l
-S
-0.000 0.804 0.000 RG
-55.59 66.26 m
-57.65 66.27 l
-59.71 66.28 l
-61.77 66.29 l
-63.83 66.31 l
-65.89 66.33 l
-67.95 66.35 l
-70.00 66.38 l
-72.06 66.41 l
-74.12 66.45 l
-76.18 66.50 l
-78.24 66.56 l
-80.30 66.63 l
-82.36 66.71 l
-84.42 66.81 l
-86.48 66.93 l
-88.54 67.07 l
-90.59 67.24 l
-92.65 67.44 l
-94.71 67.67 l
-96.77 67.94 l
-98.83 68.26 l
-100.89 68.63 l
-102.95 69.06 l
-105.01 69.56 l
-107.07 70.13 l
-109.13 70.79 l
-111.18 71.55 l
-113.24 72.41 l
-115.30 73.38 l
-117.36 74.47 l
-119.42 75.70 l
-121.48 77.07 l
-123.54 78.59 l
-125.60 80.26 l
-127.66 82.10 l
-129.72 84.11 l
-131.77 86.29 l
-133.83 88.63 l
-135.89 91.15 l
-137.95 93.84 l
-140.01 96.68 l
-142.07 99.68 l
-144.13 102.82 l
-146.19 106.08 l
-148.25 109.46 l
-150.31 112.93 l
-152.36 116.49 l
-154.42 120.09 l
-156.48 123.74 l
-158.54 127.40 l
-160.60 131.05 l
-162.66 134.68 l
-164.72 138.26 l
-166.78 141.78 l
-168.84 145.22 l
-170.90 148.56 l
-172.96 151.79 l
-175.01 154.91 l
-177.07 157.89 l
-179.13 160.75 l
-181.19 163.46 l
-183.25 166.03 l
-185.31 168.46 l
-187.37 170.75 l
-189.43 172.89 l
-191.49 174.89 l
-193.55 176.77 l
-195.60 178.51 l
-197.66 180.12 l
-199.72 181.62 l
-201.78 183.00 l
-203.84 184.28 l
-205.90 185.46 l
-207.96 186.54 l
-210.02 187.53 l
-212.08 188.44 l
-214.14 189.28 l
-216.19 190.04 l
-218.25 190.74 l
-220.31 191.38 l
-S
-Q q 301.00 60.96 177.90 142.04 re W n
-Q q 301.00 60.96 177.90 142.04 re W n
-0.000 0.000 0.000 RG
-0.75 w
-[] 0 d
-1 J
-1 j
-10.00 M
-307.59 193.86 m
-309.65 193.46 l
-311.71 193.02 l
-313.77 192.54 l
-315.83 192.02 l
-317.89 191.44 l
-319.95 190.81 l
-322.00 190.12 l
-324.06 189.36 l
-326.12 188.54 l
-328.18 187.63 l
-330.24 186.65 l
-332.30 185.58 l
-334.36 184.41 l
-336.42 183.15 l
-338.48 181.77 l
-340.54 180.29 l
-342.59 178.69 l
-344.65 176.96 l
-346.71 175.10 l
-348.77 173.11 l
-350.83 170.98 l
-352.89 168.71 l
-354.95 166.30 l
-357.01 163.74 l
-359.07 161.05 l
-361.13 158.21 l
-363.18 155.23 l
-365.24 152.13 l
-367.30 148.91 l
-369.36 145.58 l
-371.42 142.15 l
-373.48 138.64 l
-375.54 135.06 l
-377.60 131.44 l
-379.66 127.79 l
-381.72 124.13 l
-383.77 120.48 l
-385.83 116.87 l
-387.89 113.31 l
-389.95 109.83 l
-392.01 106.44 l
-394.07 103.16 l
-396.13 100.01 l
-398.19 97.00 l
-400.25 94.13 l
-402.31 91.43 l
-404.36 88.90 l
-406.42 86.53 l
-408.48 84.33 l
-410.54 82.31 l
-412.60 80.45 l
-414.66 78.76 l
-416.72 77.23 l
-418.78 75.84 l
-420.84 74.60 l
-422.90 73.49 l
-424.96 72.51 l
-427.01 71.64 l
-429.07 70.87 l
-431.13 70.20 l
-433.19 69.62 l
-435.25 69.11 l
-437.31 68.67 l
-439.37 68.29 l
-441.43 67.97 l
-443.49 67.69 l
-445.55 67.46 l
-447.60 67.26 l
-449.66 67.09 l
-451.72 66.94 l
-453.78 66.82 l
-455.84 66.72 l
-457.90 66.64 l
-459.96 66.57 l
-462.02 66.51 l
-464.08 66.46 l
-466.14 66.42 l
-468.19 66.38 l
-470.25 66.35 l
-472.31 66.33 l
-S
-Q q
-0.000 0.000 0.000 RG
-0.75 w
-[] 0 d
-1 J
-1 j
-10.00 M
-307.59 60.96 m 472.31 60.96 l S
-307.59 60.96 m 307.59 54.98 l S
-348.77 60.96 m 348.77 54.98 l S
-389.95 60.96 m 389.95 54.98 l S
-431.13 60.96 m 431.13 54.98 l S
-472.31 60.96 m 472.31 54.98 l S
-BT
-0.000 0.000 0.000 rg
-/F2 1 Tf 10.00 0.00 -0.00 10.00 301.89 39.44 Tm (-4) Tj
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 343.07 39.44 Tm (-2) Tj
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 387.17 39.44 Tm (0) Tj
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 428.35 39.44 Tm (2) Tj
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 469.53 39.44 Tm (4) Tj
-ET
-301.00 66.22 m 301.00 197.74 l S
-301.00 66.22 m 295.03 66.22 l S
-301.00 92.52 m 295.03 92.52 l S
-301.00 118.82 m 295.03 118.82 l S
-301.00 145.13 m 295.03 145.13 l S
-301.00 171.43 m 295.03 171.43 l S
-301.00 197.74 m 295.03 197.74 l S
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 286.66 59.27 Tm (0.0) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 286.66 85.57 Tm (0.2) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 286.66 111.87 Tm (0.4) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 286.66 138.18 Tm (0.6) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 286.66 164.48 Tm (0.8) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 286.66 190.79 Tm (1.0) Tj
-ET
-301.00 60.96 m
-478.90 60.96 l
-478.90 203.00 l
-301.00 203.00 l
-301.00 60.96 l
-S
-Q q 252.00 0.00 252.00 252.00 re W n
-BT
-0.000 0.000 0.000 rg
-/F3 1 Tf 12.00 0.00 -0.00 12.00 335.06 223.19 Tm [(ICC plot f) 20 (or item I4)] TJ
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 351.05 15.54 Tm (Latent Dimension) Tj
-ET
-BT
-/F2 1 Tf 0.00 10.00 -10.00 0.00 262.76 89.31 Tm [(Probability to Solv) 25 (e)] TJ
-ET
-Q q 301.00 60.96 177.90 142.04 re W n
-1.000 0.000 0.000 RG
-0.75 w
-[] 0 d
-1 J
-1 j
-10.00 M
-307.59 70.08 m
-309.65 70.47 l
-311.71 70.91 l
-313.77 71.38 l
-315.83 71.90 l
-317.89 72.47 l
-319.95 73.09 l
-322.00 73.77 l
-324.06 74.51 l
-326.12 75.33 l
-328.18 76.21 l
-330.24 77.17 l
-332.30 78.21 l
-334.36 79.35 l
-336.42 80.57 l
-338.48 81.90 l
-340.54 83.32 l
-342.59 84.86 l
-344.65 86.50 l
-346.71 88.26 l
-348.77 90.13 l
-350.83 92.12 l
-352.89 94.22 l
-354.95 96.44 l
-357.01 98.77 l
-359.07 101.19 l
-361.13 103.72 l
-363.18 106.32 l
-365.24 108.99 l
-367.30 111.72 l
-369.36 114.48 l
-371.42 117.25 l
-373.48 120.01 l
-375.54 122.73 l
-377.60 125.39 l
-379.66 127.95 l
-381.72 130.39 l
-383.77 132.67 l
-385.83 134.77 l
-387.89 136.66 l
-389.95 138.31 l
-392.01 139.70 l
-394.07 140.81 l
-396.13 141.62 l
-398.19 142.13 l
-400.25 142.31 l
-402.31 142.17 l
-404.36 141.72 l
-406.42 140.95 l
-408.48 139.89 l
-410.54 138.54 l
-412.60 136.92 l
-414.66 135.07 l
-416.72 133.00 l
-418.78 130.74 l
-420.84 128.32 l
-422.90 125.78 l
-424.96 123.13 l
-427.01 120.42 l
-429.07 117.66 l
-431.13 114.89 l
-433.19 112.13 l
-435.25 109.40 l
-437.31 106.71 l
-439.37 104.10 l
-441.43 101.56 l
-443.49 99.12 l
-445.55 96.78 l
-447.60 94.55 l
-449.66 92.43 l
-451.72 90.42 l
-453.78 88.53 l
-455.84 86.75 l
-457.90 85.09 l
-459.96 83.54 l
-462.02 82.10 l
-464.08 80.76 l
-466.14 79.52 l
-468.19 78.38 l
-470.25 77.32 l
-472.31 76.35 l
-S
-0.000 0.804 0.000 RG
-307.59 66.23 m
-309.65 66.23 l
-311.71 66.24 l
-313.77 66.24 l
-315.83 66.25 l
-317.89 66.26 l
-319.95 66.27 l
-322.00 66.28 l
-324.06 66.29 l
-326.12 66.31 l
-328.18 66.33 l
-330.24 66.35 l
-332.30 66.38 l
-334.36 66.41 l
-336.42 66.45 l
-338.48 66.50 l
-340.54 66.56 l
-342.59 66.63 l
-344.65 66.71 l
-346.71 66.81 l
-348.77 66.93 l
-350.83 67.07 l
-352.89 67.23 l
-354.95 67.43 l
-357.01 67.66 l
-359.07 67.93 l
-361.13 68.24 l
-363.18 68.61 l
-365.24 69.04 l
-367.30 69.54 l
-369.36 70.11 l
-371.42 70.77 l
-373.48 71.52 l
-375.54 72.37 l
-377.60 73.34 l
-379.66 74.43 l
-381.72 75.65 l
-383.77 77.01 l
-385.83 78.52 l
-387.89 80.19 l
-389.95 82.02 l
-392.01 84.02 l
-394.07 86.19 l
-396.13 88.53 l
-398.19 91.05 l
-400.25 93.72 l
-402.31 96.56 l
-404.36 99.55 l
-406.42 102.68 l
-408.48 105.95 l
-410.54 109.32 l
-412.60 112.79 l
-414.66 116.34 l
-416.72 119.94 l
-418.78 123.59 l
-420.84 127.25 l
-422.90 130.90 l
-424.96 134.53 l
-427.01 138.11 l
-429.07 141.63 l
-431.13 145.08 l
-433.19 148.42 l
-435.25 151.66 l
-437.31 154.78 l
-439.37 157.77 l
-441.43 160.63 l
-443.49 163.35 l
-445.55 165.93 l
-447.60 168.36 l
-449.66 170.65 l
-451.72 172.80 l
-453.78 174.81 l
-455.84 176.69 l
-457.90 178.44 l
-459.96 180.06 l
-462.02 181.56 l
-464.08 182.95 l
-466.14 184.23 l
-468.19 185.41 l
-470.25 186.50 l
-472.31 187.49 l
-S
-Q
-endstream
-endobj
-13 0 obj
-22469
-endobj
-14 0 obj
-<<
-/Type /Page
-/Parent 3 0 R
-/Contents 15 0 R
-/Resources 4 0 R
->>
-endobj
-15 0 obj
-<<
-/Length 16 0 R
->>
-stream
-1 J 1 j q
-Q q 57.60 36.00 432.00 342.00 re W n
-Q q
-0.000 0.000 0.000 RG
-0.75 w
-[] 0 d
-1 J
-1 j
-10.00 M
-57.60 36.00 m
-489.60 36.00 l
-489.60 378.00 l
-57.60 378.00 l
-57.60 36.00 l
-S
-Q q 0.00 0.00 504.00 378.00 re W n
-Q q
-0.000 0.000 0.000 RG
-0.75 w
-[] 0 d
-1 J
-1 j
-10.00 M
-57.60 112.00 m 57.60 302.00 l S
-57.60 112.00 m 50.40 112.00 l S
-57.60 175.33 m 50.40 175.33 l S
-57.60 238.67 m 50.40 238.67 l S
-57.60 302.00 m 50.40 302.00 l S
-BT
-0.000 0.000 0.000 rg
-/F2 1 Tf 8.00 0.00 -0.00 8.00 36.53 109.13 Tm (I2) Tj
-ET
-BT
-/F2 1 Tf 8.00 0.00 -0.00 8.00 36.53 172.46 Tm (I1) Tj
-ET
-BT
-/F2 1 Tf 8.00 0.00 -0.00 8.00 36.53 235.79 Tm (I4) Tj
-ET
-BT
-/F2 1 Tf 8.00 0.00 -0.00 8.00 36.53 299.13 Tm (I3) Tj
-ET
-57.60 36.00 m 489.60 36.00 l S
-98.33 36.00 m 98.33 28.80 l S
-176.94 36.00 m 176.94 28.80 l S
-255.55 36.00 m 255.55 28.80 l S
-334.16 36.00 m 334.16 28.80 l S
-412.77 36.00 m 412.77 28.80 l S
-BT
-/F2 1 Tf 8.00 0.00 -0.00 8.00 93.77 18.70 Tm (-2) Tj
-ET
-BT
-/F2 1 Tf 8.00 0.00 -0.00 8.00 172.38 18.70 Tm (-1) Tj
-ET
-BT
-/F2 1 Tf 8.00 0.00 -0.00 8.00 253.33 18.70 Tm (0) Tj
-ET
-BT
-/F2 1 Tf 8.00 0.00 -0.00 8.00 331.94 18.70 Tm (1) Tj
-ET
-BT
-/F2 1 Tf 8.00 0.00 -0.00 8.00 410.55 18.70 Tm (2) Tj
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 234.69 7.20 Tm (Latent Dimension) Tj
-ET
-Q q 57.60 36.00 432.00 342.00 re W n
-0.000 0.000 0.000 RG
-0.75 w
-[] 0 d
-1 J
-1 j
-10.00 M
-73.60 372.30 m 73.60 504.00 l S
-178.26 372.30 m 178.26 504.00 l S
-247.35 372.30 m 247.35 504.00 l S
-300.47 372.30 m 300.47 504.00 l S
-302.46 372.30 m 302.46 504.00 l S
-325.45 372.30 m 325.45 504.00 l S
-426.93 372.30 m 426.93 504.00 l S
-412.42 372.30 m 412.42 504.00 l S
-307.67 112.00 m 405.22 112.00 l S
-BT
-/F1 1 Tf 1 Tr 7.48 0 0 7.48 297.50 109.40 Tm (l) Tj 0 Tr
-ET
-BT
-/F1 1 Tf 1 Tr 7.48 0 0 7.48 409.46 109.40 Tm (l) Tj 0 Tr
-ET
-BT
-0.000 0.000 0.000 rg
-/F2 1 Tf 8.00 0.00 -0.00 8.00 298.24 100.01 Tm (1) Tj
-ET
-BT
-/F2 1 Tf 8.00 0.00 -0.00 8.00 410.20 100.01 Tm (2) Tj
-ET
-BT
-/F1 1 Tf 2 Tr 7.48 0 0 7.48 353.48 109.40 Tm (l) Tj 0 Tr
-ET
-254.55 175.33 m 419.73 175.33 l S
-BT
-/F1 1 Tf 1 Tr 7.48 0 0 7.48 244.39 172.74 Tm (l) Tj 0 Tr
-ET
-BT
-/F1 1 Tf 1 Tr 7.48 0 0 7.48 423.97 172.74 Tm (l) Tj 0 Tr
-ET
-BT
-/F2 1 Tf 8.00 0.00 -0.00 8.00 245.13 163.35 Tm (1) Tj
-ET
-BT
-/F2 1 Tf 8.00 0.00 -0.00 8.00 424.70 163.35 Tm (2) Tj
-ET
-BT
-/F1 1 Tf 2 Tr 7.48 0 0 7.48 334.18 172.74 Tm (l) Tj 0 Tr
-ET
-185.46 238.67 m 318.25 238.67 l S
-BT
-/F1 1 Tf 1 Tr 7.48 0 0 7.48 175.30 236.07 Tm (l) Tj 0 Tr
-ET
-BT
-/F1 1 Tf 1 Tr 7.48 0 0 7.48 322.49 236.07 Tm (l) Tj 0 Tr
-ET
-BT
-/F2 1 Tf 8.00 0.00 -0.00 8.00 176.03 226.68 Tm (1) Tj
-ET
-BT
-/F2 1 Tf 8.00 0.00 -0.00 8.00 323.23 226.68 Tm (2) Tj
-ET
-BT
-/F1 1 Tf 2 Tr 7.48 0 0 7.48 248.89 236.07 Tm (l) Tj 0 Tr
-ET
-80.80 302.00 m 295.26 302.00 l S
-BT
-/F1 1 Tf 1 Tr 7.48 0 0 7.48 70.64 299.40 Tm (l) Tj 0 Tr
-ET
-BT
-/F1 1 Tf 1 Tr 7.48 0 0 7.48 299.50 299.40 Tm (l) Tj 0 Tr
-ET
-BT
-/F2 1 Tf 8.00 0.00 -0.00 8.00 71.38 290.01 Tm (1) Tj
-ET
-BT
-/F2 1 Tf 8.00 0.00 -0.00 8.00 300.24 290.01 Tm (2) Tj
-ET
-BT
-/F1 1 Tf 2 Tr 7.48 0 0 7.48 185.07 299.40 Tm (l) Tj 0 Tr
-ET
-Q q
-BT
-0.000 0.000 0.000 rg
-/F2 1 Tf 12.00 0.00 -0.00 12.00 495.66 105.97 Tm ( ) Tj
-ET
-BT
-/F2 1 Tf 12.00 0.00 -0.00 12.00 495.66 169.30 Tm ( ) Tj
-ET
-BT
-/F2 1 Tf 12.00 0.00 -0.00 12.00 495.66 232.64 Tm ( ) Tj
-ET
-BT
-/F2 1 Tf 12.00 0.00 -0.00 12.00 495.66 295.97 Tm ( ) Tj
-ET
-Q q 57.60 378.00 432.00 82.80 re W n
-Q q 0.00 378.00 504.00 126.00 re W n
-BT
-0.000 0.000 0.000 rg
-/F3 1 Tf 14.00 0.00 -0.00 14.00 215.76 477.37 Tm [(P) 30 (er) 15 (son-Item Map)] TJ
-ET
-BT
-/F2 1 Tf 12.00 0.00 -0.00 12.00 267.26 323.28 Tm (ttx) Tj
-ET
-Q q 57.60 378.00 432.00 82.80 re W n
-0.745 0.745 0.745 RG
-3.75 w
-[] 0 d
-2 J
-1 j
-10.00 M
-83.30 381.07 m 83.30 384.90 l S
-173.08 381.07 m 173.08 403.11 l S
-236.61 381.07 m 236.61 414.61 l S
-290.24 381.07 m 290.24 457.73 l S
-341.63 381.07 m 341.63 414.61 l S
-397.79 381.07 m 397.79 429.94 l S
-473.60 381.07 m 473.60 407.90 l S
-Q q
-BT
-0.000 0.000 0.000 rg
-/F2 1 Tf 8.00 0.00 -0.00 8.00 25.46 426.61 Tm [(P) 50 (erson)] TJ
-ET
-BT
-/F2 1 Tf 8.00 0.00 -0.00 8.00 13.46 416.53 Tm [(P) 40 (ar) 10 (ameter)] TJ
-ET
-BT
-/F2 1 Tf 8.00 0.00 -0.00 8.00 10.43 406.45 Tm [(Distr) -15 (ib) 20 (ution)] TJ
-ET
-0.000 0.000 0.000 RG
-0.75 w
-[] 0 d
-1 J
-1 j
-10.00 M
-57.60 378.00 m
-489.60 378.00 l
-489.60 460.80 l
-57.60 460.80 l
-57.60 378.00 l
-S
-Q q 59.04 73.44 414.72 371.52 re W n
-Q
-endstream
-endobj
-16 0 obj
-4275
-endobj
-17 0 obj
-<<
-/Type /Page
-/Parent 3 0 R
-/Contents 18 0 R
-/Resources 4 0 R
->>
-endobj
-18 0 obj
-<<
-/Length 19 0 R
->>
-stream
-1 J 1 j q
-Q q 57.60 36.00 432.00 342.00 re W n
-Q q
-0.000 0.000 0.000 RG
-0.75 w
-[] 0 d
-1 J
-1 j
-10.00 M
-57.60 36.00 m
-489.60 36.00 l
-489.60 378.00 l
-57.60 378.00 l
-57.60 36.00 l
-S
-Q q 0.00 0.00 504.00 378.00 re W n
-Q q
-0.000 0.000 0.000 RG
-0.75 w
-[] 0 d
-1 J
-1 j
-10.00 M
-57.60 112.00 m 57.60 302.00 l S
-57.60 112.00 m 50.40 112.00 l S
-57.60 175.33 m 50.40 175.33 l S
-57.60 238.67 m 50.40 238.67 l S
-57.60 302.00 m 50.40 302.00 l S
-BT
-0.000 0.000 0.000 rg
-/F2 1 Tf 8.00 0.00 -0.00 8.00 36.53 109.13 Tm (I2) Tj
-ET
-BT
-/F2 1 Tf 8.00 0.00 -0.00 8.00 36.53 172.46 Tm (I1) Tj
-ET
-BT
-/F2 1 Tf 8.00 0.00 -0.00 8.00 36.53 235.79 Tm (I4) Tj
-ET
-BT
-/F2 1 Tf 8.00 0.00 -0.00 8.00 36.53 299.13 Tm (I3) Tj
-ET
-57.60 36.00 m 489.60 36.00 l S
-98.33 36.00 m 98.33 28.80 l S
-176.94 36.00 m 176.94 28.80 l S
-255.55 36.00 m 255.55 28.80 l S
-334.16 36.00 m 334.16 28.80 l S
-412.77 36.00 m 412.77 28.80 l S
-BT
-/F2 1 Tf 8.00 0.00 -0.00 8.00 93.77 18.70 Tm (-2) Tj
-ET
-BT
-/F2 1 Tf 8.00 0.00 -0.00 8.00 172.38 18.70 Tm (-1) Tj
-ET
-BT
-/F2 1 Tf 8.00 0.00 -0.00 8.00 253.33 18.70 Tm (0) Tj
-ET
-BT
-/F2 1 Tf 8.00 0.00 -0.00 8.00 331.94 18.70 Tm (1) Tj
-ET
-BT
-/F2 1 Tf 8.00 0.00 -0.00 8.00 410.55 18.70 Tm (2) Tj
-ET
-BT
-/F2 1 Tf 10.00 0.00 -0.00 10.00 234.69 7.20 Tm (Latent Dimension) Tj
-ET
-Q q 57.60 36.00 432.00 342.00 re W n
-0.000 0.000 0.000 RG
-0.75 w
-[] 0 d
-1 J
-1 j
-10.00 M
-73.60 372.30 m 73.60 504.00 l S
-178.26 372.30 m 178.26 504.00 l S
-247.35 372.30 m 247.35 504.00 l S
-300.47 372.30 m 300.47 504.00 l S
-302.46 372.30 m 302.46 504.00 l S
-325.45 372.30 m 325.45 504.00 l S
-426.93 372.30 m 426.93 504.00 l S
-412.42 372.30 m 412.42 504.00 l S
-307.67 112.00 m 405.22 112.00 l S
-BT
-/F1 1 Tf 1 Tr 7.48 0 0 7.48 297.50 109.40 Tm (l) Tj 0 Tr
-ET
-BT
-/F1 1 Tf 1 Tr 7.48 0 0 7.48 409.46 109.40 Tm (l) Tj 0 Tr
-ET
-BT
-0.000 0.000 0.000 rg
-/F2 1 Tf 8.00 0.00 -0.00 8.00 298.24 100.01 Tm (1) Tj
-ET
-BT
-/F2 1 Tf 8.00 0.00 -0.00 8.00 410.20 100.01 Tm (2) Tj
-ET
-BT
-/F1 1 Tf 2 Tr 7.48 0 0 7.48 353.48 109.40 Tm (l) Tj 0 Tr
-ET
-254.55 175.33 m 419.73 175.33 l S
-BT
-/F1 1 Tf 1 Tr 7.48 0 0 7.48 244.39 172.74 Tm (l) Tj 0 Tr
-ET
-BT
-/F1 1 Tf 1 Tr 7.48 0 0 7.48 423.97 172.74 Tm (l) Tj 0 Tr
-ET
-BT
-/F2 1 Tf 8.00 0.00 -0.00 8.00 245.13 163.35 Tm (1) Tj
-ET
-BT
-/F2 1 Tf 8.00 0.00 -0.00 8.00 424.70 163.35 Tm (2) Tj
-ET
-BT
-/F1 1 Tf 2 Tr 7.48 0 0 7.48 334.18 172.74 Tm (l) Tj 0 Tr
-ET
-185.46 238.67 m 318.25 238.67 l S
-BT
-/F1 1 Tf 1 Tr 7.48 0 0 7.48 175.30 236.07 Tm (l) Tj 0 Tr
-ET
-BT
-/F1 1 Tf 1 Tr 7.48 0 0 7.48 322.49 236.07 Tm (l) Tj 0 Tr
-ET
-BT
-/F2 1 Tf 8.00 0.00 -0.00 8.00 176.03 226.68 Tm (1) Tj
-ET
-BT
-/F2 1 Tf 8.00 0.00 -0.00 8.00 323.23 226.68 Tm (2) Tj
-ET
-BT
-/F1 1 Tf 2 Tr 7.48 0 0 7.48 248.89 236.07 Tm (l) Tj 0 Tr
-ET
-80.80 302.00 m 295.26 302.00 l S
-BT
-/F1 1 Tf 1 Tr 7.48 0 0 7.48 70.64 299.40 Tm (l) Tj 0 Tr
-ET
-BT
-/F1 1 Tf 1 Tr 7.48 0 0 7.48 299.50 299.40 Tm (l) Tj 0 Tr
-ET
-BT
-/F2 1 Tf 8.00 0.00 -0.00 8.00 71.38 290.01 Tm (1) Tj
-ET
-BT
-/F2 1 Tf 8.00 0.00 -0.00 8.00 300.24 290.01 Tm (2) Tj
-ET
-BT
-/F1 1 Tf 2 Tr 7.48 0 0 7.48 185.07 299.40 Tm (l) Tj 0 Tr
-ET
-Q q
-BT
-0.000 0.000 0.000 rg
-/F2 1 Tf 12.00 0.00 -0.00 12.00 495.66 105.97 Tm ( ) Tj
-ET
-BT
-/F2 1 Tf 12.00 0.00 -0.00 12.00 495.66 169.30 Tm ( ) Tj
-ET
-BT
-/F2 1 Tf 12.00 0.00 -0.00 12.00 495.66 232.64 Tm ( ) Tj
-ET
-BT
-/F2 1 Tf 12.00 0.00 -0.00 12.00 495.66 295.97 Tm ( ) Tj
-ET
-Q q 57.60 378.00 432.00 82.80 re W n
-Q q 0.00 378.00 504.00 126.00 re W n
-BT
-0.000 0.000 0.000 rg
-/F3 1 Tf 14.00 0.00 -0.00 14.00 215.76 477.37 Tm [(P) 30 (er) 15 (son-Item Map)] TJ
-ET
-BT
-/F2 1 Tf 12.00 0.00 -0.00 12.00 267.26 323.28 Tm (ttx) Tj
-ET
-Q q 57.60 378.00 432.00 82.80 re W n
-0.745 0.745 0.745 RG
-3.75 w
-[] 0 d
-2 J
-1 j
-10.00 M
-83.30 381.07 m 83.30 384.90 l S
-173.08 381.07 m 173.08 403.11 l S
-236.61 381.07 m 236.61 414.61 l S
-290.24 381.07 m 290.24 457.73 l S
-341.63 381.07 m 341.63 414.61 l S
-397.79 381.07 m 397.79 429.94 l S
-473.60 381.07 m 473.60 407.90 l S
-Q q
-BT
-0.000 0.000 0.000 rg
-/F2 1 Tf 8.00 0.00 -0.00 8.00 25.46 426.61 Tm [(P) 50 (erson)] TJ
-ET
-BT
-/F2 1 Tf 8.00 0.00 -0.00 8.00 13.46 416.53 Tm [(P) 40 (ar) 10 (ameter)] TJ
-ET
-BT
-/F2 1 Tf 8.00 0.00 -0.00 8.00 10.43 406.45 Tm [(Distr) -15 (ib) 20 (ution)] TJ
-ET
-0.000 0.000 0.000 RG
-0.75 w
-[] 0 d
-1 J
-1 j
-10.00 M
-57.60 378.00 m
-489.60 378.00 l
-489.60 460.80 l
-57.60 460.80 l
-57.60 378.00 l
-S
-Q q 59.04 73.44 414.72 371.52 re W n
-Q
-endstream
-endobj
-19 0 obj
-4275
-endobj
-3 0 obj
-<<
-/Type /Pages
-/Kids [
-5 0 R
-8 0 R
-11 0 R
-14 0 R
-17 0 R
-]
-/Count 5
-/MediaBox [0 0 504 504]
->>
-endobj
-4 0 obj
-<<
-/ProcSet [/PDF /Text]
-/Font << /F1 21 0 R /F2 22 0 R /F3 23 0 R >>
-/ExtGState << >>
->>
-endobj
-20 0 obj
-<<
-/Type /Encoding
-/BaseEncoding /WinAnsiEncoding
-/Differences [ 45/minus ]
->>
-endobj
-21 0 obj
-<<
-/Type /Font
-/Subtype /Type1
-/Name /F1
-/BaseFont /ZapfDingbats
->>
-endobj
-22 0 obj <<
-/Type /Font
-/Subtype /Type1
-/Name /F2
-/BaseFont /Helvetica
-/Encoding 20 0 R
->> endobj
-23 0 obj <<
-/Type /Font
-/Subtype /Type1
-/Name /F3
-/BaseFont /Helvetica-Bold
-/Encoding 20 0 R
->> endobj
-xref
-0 24
-0000000000 65535 f
-0000000021 00000 n
-0000000164 00000 n
-0000074182 00000 n
-0000074292 00000 n
-0000000213 00000 n
-0000000293 00000 n
-0000020042 00000 n
-0000020063 00000 n
-0000020143 00000 n
-0000042666 00000 n
-0000042688 00000 n
-0000042770 00000 n
-0000065294 00000 n
-0000065316 00000 n
-0000065398 00000 n
-0000069728 00000 n
-0000069749 00000 n
-0000069831 00000 n
-0000074161 00000 n
-0000074397 00000 n
-0000074492 00000 n
-0000074576 00000 n
-0000074674 00000 n
-trailer
-<<
-/Size 24
-/Info 1 0 R
-/Root 2 0 R
->>
-startxref
-74777
-%%EOF
diff --git a/inst/doc/UCML.jpg b/inst/doc/UCML.jpg
old mode 100644
new mode 100755
diff --git a/inst/doc/Z.cls b/inst/doc/Z.cls
old mode 100644
new mode 100755
diff --git a/inst/doc/eRm.R b/inst/doc/eRm.R
deleted file mode 100644
index 3beec56..0000000
--- a/inst/doc/eRm.R
+++ /dev/null
@@ -1,101 +0,0 @@
-###################################################
-### chunk number 1:
-###################################################
-library(eRm)
-data(raschdat1)
-res.rasch <- RM(raschdat1)
-pres.rasch <- person.parameter(res.rasch)
-
-
-###################################################
-### chunk number 2:
-###################################################
-lrres.rasch <- LRtest(res.rasch, splitcr = "mean", se = TRUE)
-lrres.rasch
-
-
-###################################################
-### chunk number 3:
-###################################################
-plotGOF(lrres.rasch, beta.subset=c(14,5,18,7,1), tlab="item", conf=list(ia=FALSE,col="blue",lty="dotted"))
-
-
-###################################################
-### chunk number 4:
-###################################################
-data(lltmdat2)
-W <- matrix(c(1,2,1,3,2,2,2,1,1,1),ncol=2)
-res.lltm <- LLTM(lltmdat2, W)
-summary(res.lltm)
-
-
-###################################################
-### chunk number 5:
-###################################################
-data(pcmdat2)
-res.rsm <- RSM(pcmdat2)
-thresholds(res.rsm)
-
-
-###################################################
-### chunk number 6:
-###################################################
-plotICC(res.rsm, mplot=TRUE, legpos=FALSE,ask=FALSE)
-
-
-###################################################
-### chunk number 7:
-###################################################
-plotICC(res.rsm, mplot=TRUE, legpos=FALSE,ask=FALSE)
-
-
-###################################################
-### chunk number 8:
-###################################################
-res.pcm <- PCM(pcmdat2)
-plotPImap(res.pcm, sorted = TRUE)
-
-
-###################################################
-### chunk number 9:
-###################################################
-res.pcm <- PCM(pcmdat2)
-plotPImap(res.pcm, sorted = TRUE)
-
-
-###################################################
-### chunk number 10:
-###################################################
-pres.pcm <- person.parameter(res.pcm)
-itemfit(pres.pcm)
-
-
-###################################################
-### chunk number 11:
-###################################################
-lr<- 2*(res.pcm$loglik-res.rsm$loglik)
-df<- res.pcm$npar-res.rsm$npar
-pvalue<-1-pchisq(lr,df)
-cat("LR statistic: ", lr, " df =",df, " p =",pvalue, "\n")
-
-
-###################################################
-### chunk number 12:
-###################################################
-data(lpcmdat)
-grouplpcm <- rep(1:2, each = 10)
-
-
-###################################################
-### chunk number 13:
-###################################################
-reslpcm <- LPCM(lpcmdat, mpoints = 2, groupvec = grouplpcm, sum0 = FALSE)
-model.matrix(reslpcm)
-
-
-###################################################
-### chunk number 14:
-###################################################
-reslpcm
-
-
diff --git a/inst/doc/eRm.Rnw b/inst/doc/eRm.Rnw
old mode 100644
new mode 100755
index 68afe9f..5ee2398
--- a/inst/doc/eRm.Rnw
+++ b/inst/doc/eRm.Rnw
@@ -1002,5 +1002,8 @@ person performances in tests. This improves the feasibility of IRT
models with respect to a wide variety of application areas.
\bibliography{eRmvig}
+\newpage
+
+\rotatebox[origin=c]{90}{\includegraphics[width=1.1\textheight]{eRm_object_tree.pdf}}
\end{document}
diff --git a/inst/doc/eRm.pdf b/inst/doc/eRm.pdf
index 0388b5d..1c389ee 100644
Binary files a/inst/doc/eRm.pdf and b/inst/doc/eRm.pdf differ
diff --git a/inst/doc/eRm_object_tree.pdf b/inst/doc/eRm_object_tree.pdf
new file mode 100755
index 0000000..7baa54a
Binary files /dev/null and b/inst/doc/eRm_object_tree.pdf differ
diff --git a/inst/doc/eRmvig.R b/inst/doc/eRmvig.R
deleted file mode 100644
index 3beec56..0000000
--- a/inst/doc/eRmvig.R
+++ /dev/null
@@ -1,101 +0,0 @@
-###################################################
-### chunk number 1:
-###################################################
-library(eRm)
-data(raschdat1)
-res.rasch <- RM(raschdat1)
-pres.rasch <- person.parameter(res.rasch)
-
-
-###################################################
-### chunk number 2:
-###################################################
-lrres.rasch <- LRtest(res.rasch, splitcr = "mean", se = TRUE)
-lrres.rasch
-
-
-###################################################
-### chunk number 3:
-###################################################
-plotGOF(lrres.rasch, beta.subset=c(14,5,18,7,1), tlab="item", conf=list(ia=FALSE,col="blue",lty="dotted"))
-
-
-###################################################
-### chunk number 4:
-###################################################
-data(lltmdat2)
-W <- matrix(c(1,2,1,3,2,2,2,1,1,1),ncol=2)
-res.lltm <- LLTM(lltmdat2, W)
-summary(res.lltm)
-
-
-###################################################
-### chunk number 5:
-###################################################
-data(pcmdat2)
-res.rsm <- RSM(pcmdat2)
-thresholds(res.rsm)
-
-
-###################################################
-### chunk number 6:
-###################################################
-plotICC(res.rsm, mplot=TRUE, legpos=FALSE,ask=FALSE)
-
-
-###################################################
-### chunk number 7:
-###################################################
-plotICC(res.rsm, mplot=TRUE, legpos=FALSE,ask=FALSE)
-
-
-###################################################
-### chunk number 8:
-###################################################
-res.pcm <- PCM(pcmdat2)
-plotPImap(res.pcm, sorted = TRUE)
-
-
-###################################################
-### chunk number 9:
-###################################################
-res.pcm <- PCM(pcmdat2)
-plotPImap(res.pcm, sorted = TRUE)
-
-
-###################################################
-### chunk number 10:
-###################################################
-pres.pcm <- person.parameter(res.pcm)
-itemfit(pres.pcm)
-
-
-###################################################
-### chunk number 11:
-###################################################
-lr<- 2*(res.pcm$loglik-res.rsm$loglik)
-df<- res.pcm$npar-res.rsm$npar
-pvalue<-1-pchisq(lr,df)
-cat("LR statistic: ", lr, " df =",df, " p =",pvalue, "\n")
-
-
-###################################################
-### chunk number 12:
-###################################################
-data(lpcmdat)
-grouplpcm <- rep(1:2, each = 10)
-
-
-###################################################
-### chunk number 13:
-###################################################
-reslpcm <- LPCM(lpcmdat, mpoints = 2, groupvec = grouplpcm, sum0 = FALSE)
-model.matrix(reslpcm)
-
-
-###################################################
-### chunk number 14:
-###################################################
-reslpcm
-
-
diff --git a/inst/doc/eRmvig.Rnw b/inst/doc/eRmvig.Rnw
deleted file mode 100644
index 68afe9f..0000000
--- a/inst/doc/eRmvig.Rnw
+++ /dev/null
@@ -1,1006 +0,0 @@
-%\VignetteIndexEntry{eRm Basics}
-
-\documentclass[article]{Z}
-\usepackage{amsmath, thumbpdf}
-\usepackage{Sweave}
-\usepackage{graphicx}
-
-\author{Patrick Mair\\Wirtschaftsuniversit\"at Wien \And
- Reinhold Hatzinger\\Wirtschaftsuniversit\"at Wien}
-\Plainauthor{Patrick Mair, Reinhold Hatzinger}
-
-\title{Extended Rasch Modeling: The R Package \pkg{eRm}}
-\Plaintitle{Extended Rasch Modeling: The R Package eRm}
-\Shorttitle{The R Package \pkg{eRm}}
-
-\Abstract{
-
-This package vignette is an update of the \pkg{eRm} papers by published
-in a special issue on Psychometrics in the Journal of Statistical
-Software and in Psychology Science \citep{Mair+Hatzinger:2007,
-Mair+Hatzinger:2007b}. Since the publication of these papers various
-extensions and additional features have been incorporated into the
-package. We start with a methodological introduction to extended
-Rasch models followed by a general program description and application
-topics. The package allows for the computation of simple Rasch models,
-rating scale models, partial credit models and linear extensions of
-these. The incorporation of such linear structures allows for modeling
-the effects of covariates and enables the analysis of repeated
-categorical measurements. The item parameter estimation is
-performed by means of CML, for the person parameters we use ordinary ML.
-The estimation routines work for incomplete data matrices as well.
-Based on these estimators, item-wise and global goodness-of-fit
-statistics are described and various plots are presented. }
-
-\Keywords{eRm package, Rasch model, LLTM, RSM, LRSM, PCM, LPCM, CML estimation}
-
-%\Volume{20}
-%\Issue{9}
-%\Month{April}
-%\Year{2007}
-%FIXME%
-%% \Submitdate{2004-06-21}
-%% \Acceptdate{2004-12-04}
-
-%\Address{
-% Patrick Mair\\
-% Department f\"ur Statistik und Mathematik\\
-% Wirtschaftsuniversit\"at Wien\\
-% A-1090 Wien, Austria\\
-% E-mail: \email{patrick.mair at wu-wien.ac.at}\\
-% URL: \url{http://statmath.wu-wien.ac.at/~mair/}
-%}
-
-\begin{document}
-
-\section{Introduction}
-
-\citet{Ro:99} claimed in his article that ``even though the Rasch model
-has been existing for such a long time, 95\% of the current tests in
-psychology are still constructed by using methods from classical test
-theory" (p. 140). Basically, he quotes the following reasons why the
-Rasch model (RM) is being rarely used: The Rasch model in its original form
-\citep{Ra:60}, which was limited to dichotomous items, is arguably too
-restrictive for practical testing purposes. Thus, researchers should
-focus on extended Rasch models. In addition, Rost argues that there is
-a lack of user-friendly software for the computation of such models.
-Hence, there is a need for a comprehensive, user-friendly software
-routine. Corresponding recent discussions can be found in
-\citet{Kub:05} and \citet{Bor:06}.
-
-In addition to the RM, the models that can be computed by means of the \pkg {eRm} package are:
-the linear logistic test model \citep{Scheib:72}, the rating scale model
-\citep{And:78}, the linear rating scale model \citep{FiPa:91}, the
-partial credit model \citep{Mast:82}, and the linear partial credit
-model \citep{GlVe:89,FiPo:94}. These models and their main
-characteristics are presented in Section \ref{sec:erm}.
-
-Concerning parameter estimation, these models have an important feature
-in common: Separability of item and person parameters. This implies
-that the item parameters $\mathbf{\beta}$ can be estimated without
-estimating the person parameters achieved by conditioning the likelihood
-on the sufficient person raw score. This conditional maximum likelihood
-(CML) approach is described in Section \ref{sec:cml}.
-
-Several diagnostic tools and tests to evaluate model fit are presented in Section \ref{Gof}.
-
-In Section \ref{sec:pack}, the corresponding implementation in
-\proglang{R} \citep{R:06} is described by means of several examples.
-The \pkg{eRm} package uses a design matrix approach which allows
-to reparameterize the item parameters to model common characteristics of
-the items or to enable the
-user to impose repeated measurement designs as well as group contrasts.
-By combining these types of contrasts one allows that the item parameter
-may differ over time with respect to certain subgroups. To illustrate
-the flexibility of the \pkg{eRm} package some examples are given to show
-how suitable design matrices can be constructed.
-
-%----------------- end introduction ----------------
-\section{Extended Rasch models}
-\label{sec:erm}
-
-\subsection{General expressions}
-Briefly after the first publication of the basic Rasch Model \citep{Ra:60}, the author worked on polytomous generalizations which can be found in \citet{Ra:61}. \citet{And:95} derived the representations below which are based on Rasch's general expression for polytomous data. The data matrix is denoted as $\mathbf{X}$ with the persons in the rows and the items in the columns. In total there are $v=1,...,n$ persons and $i=1,...,k$ items. A single element in the data matrix $\mathbf{X}$ is [...]
-
-\begin{equation}
-\label{eq1}
- P(X_{vi}=h)=\frac{\exp[\phi_h(\theta_v+\beta_i)+\omega_h]}{\sum_{l=0}^{m_i} \exp[\phi_l (\theta_v+\beta_i)+\omega_l]}
-\end{equation}
-
-or
-
-\begin{equation}
-\label{eq2}
- P(X_{vi}=h)=\frac{\exp[\phi_h \theta_v+\beta_{ih}]}{\sum_{l=0}^{m_i} \exp[\phi_l \theta_v+\beta_{il}]}.
-\end{equation}
-
-Here, $\phi_h$ are scoring functions for the item parameters, $\theta_v$ are the uni-dimensional person parameters, and $\beta_i$ are the item parameters. In Equation \ref{eq1}, $\omega_h$ corresponds to category parameters, whereas in Equation \ref{eq2} $\beta_{ih}$ are the item-category parameters. The meaning of these parameters will be discussed in detail below. Within the framework of these two equations, numerous models have been suggested that retain the basic properties of the Ra [...]
-
-
-\subsection{Representation of extended Rasch models}
-\label{Rep}
-For the ordinary Rasch model for dichotomous items, Equation \ref{eq1} reduces to
-\begin{equation}
-\label{eq:rasch}
- P(X_{vi}=1)=\frac{\exp(\theta_v - \beta_i)}{1+\exp(\theta_v-\beta_i)}.
-\end{equation}
-The main assumptions, which hold as well for the generalizations presented in this paper, are: uni-dimensionality of the latent trait, sufficiency of the raw score, local independence, and parallel item characteristic curves (ICCs). Corresponding explanations can be found, e.g., in \citet{Fisch:74} and mathematical derivations and proofs in \citet{Fisch:95a}.
-
-\begin{figure}[hbt]
-\centering
-\includegraphics[height=60mm, width=40mm]{modelhierarchy.pdf}
-\caption{\label{fig1} Model hierarchy}
-\end{figure}
-
-For dichotomous items, \citet{Scheib:72} proposed the (even more restricted) linear logistic test model (LLTM), later formalized by \citet{Fisch:73}, by splitting up the item parameters into the linear combination
-
-\begin{equation}
-\label{eq4}
- \beta_i=\sum_{j=1}^p w_{ij} \eta_j.
-\end{equation}
-
-\citet{Scheib:72} explained the dissolving process of items in a test for logics (``Mengenrechentest") by so-called ``cognitive operations" $\eta_j$ such as negation, disjunction, conjunction, sequence, intermediate result, permutation, and material. Note that the weights $w_{ij}$ for item $i$ and operation $j$ have to be fixed a priori. Further elaborations about the cognitive operations can be found in \citet[p.~361ff.]{Fisch:74}. Thus, from this perspective the LLTM is more parsimonou [...]
-
-Though, there exists another way to look at the LLTM: A generalization of the basic Rasch model in terms of repeated measures and group contrasts. It should be noted that both types of reparameterization also apply to the linear rating scale model (LRSM) and the linear partial credit model (LPCM) with respect to the basic rating scale model (RSM) and the partial credit model (PCM) presented below. Concerning the LLTM, the possibility to use it as a generalization of the Rasch model for r [...]
-
-At this point we will focus on a simple polytomous generalization of the Rasch model, the RSM \citep{And:78}, where each item $I_i$ must have the same number of categories. Pertaining to Equation \ref{eq1}, $\phi_h$ may be set to $h$ with $h=0,...,m$. Since in the RSM the number of item categories is constant, $m$ is used instead of $m_i$. Hence, it follows that
-
-\begin{equation}
-\label{eq5}
- P(X_{vi}=h)=\frac{\exp[h(\theta_v+\beta_i)+\omega_h]}{\sum_{l=0}^m \exp[l(\theta_v+ \beta_i)+\omega_l]},
-\end{equation}
-
-with $k$ item parameters $\beta_1,...,\beta_k$ and $m+1$ category parameters $\omega_0,...,\omega_m$. This parameterization causes a scoring of the response categories $C_h$ which is constant over the single items. Again, the item parameters can be split up in a linear combination as in Equation \ref{eq4}. This leads to the LRSM proposed by \citet{FiPa:91}.
-
-Finally, the PCM developed by \citet{Mast:82} and its linear extension, the LPCM \citep{FiPo:94}, are presented. The PCM assigns one parameter $\beta_{ih}$ to each $I_i \times C_h$ combination for $h=0,...,m_i$. Thus, the constant scoring property must not hold over the items and in addition, the items can have different numbers of response categories denoted by $m_i$. Therefore, the PCM can be regarded as a generalization of the RSM and the probability for a response of person $v$ on ca [...]
-
-\begin{equation}
-\label{eq6}
- P(X_{vih}=1)=\frac{\exp[h\theta_v + \beta_{ih}]}{\sum_{l=0}^{m_i}\exp[l\theta_v + \beta_{il}]}.
-\end{equation}
-
-It is obvious that (\ref{eq6}) is a simplification of (\ref{eq2}) in terms of $\phi_h = h$. As for the LLTM and the LRSM, the LPCM is defined by reparameterizing the item parameters of the basic model, i.e.,
-
-\begin{equation}
-\label{eq:lpcmeta}
- \beta_{ih}=\sum_{j=1}^p w_{ihj}\eta_j.
-\end{equation}
-
-These six models constitute a hierarchical order as displayed in Figure \ref{fig1}. This hierarchy is the base for a unified CML approach presented in the next section. It is outlined again that the linear extension models can be regarded either as generalizations or as more restrictive formulations pertaining to the underlying base model. The hierarchy for the basic model is straightforward: The RM allows only items with two categories, thus each item is represented by one parameter $\b [...]
-
-To conclude, the most general model is the LPCM. All other models can be considered as simplifications of Equation \ref{eq6} combined with Equation \ref{eq:lpcmeta}. As a consequence, once an estimation procedure is established for the LPCM, this approach can be used for any of the remaining models. This is what we quote as \textit{unified CML approach}. The corresponding likelihood equations follow in Section \ref{sec:cml}.
-
-\subsection{The concept of virtual items}
-\label{sec:design}
-When operating with longitudinal models, the
-main research question is whether an individual's test
-performance changes over time. The most intuitive way would be to
-look at the shift in ability $\theta_v$ across time points. Such
-models are presented e.g. in \citet{Mi:85}, \citet{Glas:1992}, and
-discussed by \citet{Ho:95}.
-
-Yet there exists another look onto time dependent changes, as presented in \citet[p~158ff.]{Fisch:95b}: The
-person parameters are fixed over time and instead of them the item
-parameters change. The basic idea is that one item $I_i$ is presented at two different times to the same person $S_v$
-is regarded as a pair of \textit{virtual items}. Within the framework of extended Rasch models, any change in $\theta_v$ occuring between the testing occasions can be described without loss of generality as a change of the item parameters, instead of describing change in terms of the person parameter. Thus, with only two measurement points, $I_i$ with the corresponding parameter $\beta_i$ generates two virtual items $I_r$ and $I_s$ with associated item parameters $\beta^{\ast}_r$ and $\b [...]
-
-Correspondingly, for each measurement point $t$ we have a vector of
-\textit{virtual item parameters} $\boldsymbol{\beta}^{\ast(t)}$ of
-length $k$. These are linear reparameterizations of the original
-$\boldsymbol{\beta}^{(t)}$, and thus the CML approach can be used
-for estimation. In general, for a simple LLTM with two measurement points the design
-matrix $\boldsymbol{W}$ is of the form as given in Table \ref{tab1}.
-
-\begin{table}
-\centering
-\[
-\begin{array}{c|c|rrrr|r}
-& & \eta_1 & \eta_2 & \hdots & \eta_k & \eta_{k+1}\\
-\hline
-\textrm{Time 1} & \beta_1^{\ast(1)} & 1 & 0 & 0 & 0 & 0\\
-& \beta_2^{\ast(1)} & 0 & 1 & 0 & 0 & 0\\
-& \vdots & & & \ddots& & \vdots\\
-& \beta_{k}^{\ast(1)} & 1 & 0 & 0 & 1 & 0\\
-\hline
-\textrm{Time 2} & \beta_{k+1}^{\ast(2)} & 1 & 0 & 0 & 0 & 1\\
-& \beta_{k+2}^{\ast(2)} & 0 & 1 & 0 & 0 & 1\\
-& \vdots & & & \ddots& & \vdots\\
-& \beta_{2k}^{\ast(2)} & 1 & 0 & 0 & 1 & 1\\
-\end{array}
-\]
-\caption{\label{tab1}A design matrix for an LLTM with two timepoints.}
-\end{table}
-
-The parameter vector $\boldsymbol{\beta}^{\ast(1)}$ represents the
-item parameters for the first test occasion,
-$\boldsymbol{\beta}^{\ast(2)}$ the parameters for the second
-occasion. It might be of interest whether these vectors differ. The
-corresponding trend contrast is $\eta_{k+1}$. Due to this contrast,
-the number of original $\beta$-parameters is doubled by introducing
-the $2k$ virtual item parameters. If we assume a constant shift for
-all item parameters, it is only necessary to estimate
-$\hat{\boldsymbol{\eta}}'=(\hat{\eta}_1,...,\hat{\eta}_{k+1})$
-where $\hat{\eta}_{k+1}$ gives the amount of shift. Since according to (\ref{eq4}), the vector
-$\hat{\boldsymbol{\beta}}^\ast$ is just a linear combination of
-$\hat{\boldsymbol{\eta}}$.
-
-As mentioned in the former section, when using models with linear
-extensions it is possible to impose group contrasts. By doing this,
-one allows that the item difficulties are different across
-subgroups. However, this is possible only for models with repeated
-measurements and virtual items since otherwise the introduction of a
-group contrast leads to overparameterization and the group effect
-cannot be estimated by using CML.
-
-Table \ref{tab2} gives an example for a repeated measurement design
-where the effect of a treatment is to be evaluated by comparing item
-difficulties regarding a control and a treatment group. The number
-of virtual parameters is doubled compared to the model matrix given
-in Table \ref{tab1}.
-
-\begin{table}[h]
- \centering
-\[
-\begin{array}{c|c|c|rrrr|rrr}
-& & & \eta_1 & \eta_2 & \hdots & \eta_k & \eta_{k+1} & \eta_{k+2} \\
-\hline
-\textrm{Time 1} & \textrm{Group 1} & \beta_1^{\ast(1)} & 1 & 0 & 0 & 0 & 0 & 0\\
-& & \beta_2^{\ast(1)} & 0 & 1 & 0 & 0 & 0& 0\\
-& & \vdots & & & \ddots& &\vdots &\vdots\\
-& & \beta_{k}^{\ast(1)} & 1 & 0 & 0 & 1 & 0 & 0\\
-\cline{2-9}
-& \textrm{Group 2} & \beta_{k+1}^{\ast(1)} & 1 & 0 & 0 & 0 & 0 & 0\\
-& & \beta_{k+2}^{\ast(1)} & 0 & 1 & 0 & 0 & 0 & 0\\
-& & \vdots & & & \ddots& &\vdots & \vdots\\
-& & \beta_{2k}^{\ast(1)} & 1 & 0 & 0 & 1 & 0& 0\\
-\hline
-\textrm{Time 2} & \textrm{Group 1} & \beta_1^{\ast(2)} & 1 & 0 & 0 & 0 & 1 & 0\\
-& & \beta_2^{\ast(2)} & 0 & 1 & 0 & 0 & 1 & 0\\
-& & \vdots & & & \ddots& &\vdots &\vdots\\
-& & \beta_{k}^{\ast(2)} & 1 & 0 & 0 & 1 & 1 & 0\\
-\cline{2-9}
-& \textrm{Group 2} & \beta_{k+1}^{\ast(2)} & 1 & 0 & 0 & 0 & 1 & 1\\
-& & \beta_{k+2}^{\ast(2)} & 0 & 1 & 0 & 0 & 1 & 1\\
-& & \vdots & & & \ddots& &\vdots & \vdots\\
-& & \beta_{2k}^{\ast(2)} & 1 & 0 & 0 & 1 & 1 & 1\\
-\end{array} \]
-\caption{\label{tab2} Design matrix for a repeated measurements design with treatment and control group.}
-\end{table}
-
-Again, $\eta_{k+1}$ is the parameter that refers to the time
-contrast, and $\eta_{k+2}$ is a group effect within
-measurement point 2. More examples are given in Section \ref{sec:pack}
-and further explanations can be found in \citet{Fisch:95b},
-\citet{FiPo:94}, and in the software manual for the LPCM-Win program
-by \citet{FiPS:98}.
-
-By introducing the concept of virtual persons, \pkg{eRm} allows for the computation of the linear logistic test model with relaxed assumptions \citep[LLRA][]{Fisch:77}. Corresponding explanations will be given in a subsequent version of this vignette.
-
-
-%------------------------ end extended Rasch models --------------------------
-
-\section{Estimation of item and person parameters}
-\label{sec:cml}
-
-\subsection{CML for item parameter estimation}
-The main idea behind the CML estimation is that the person's raw score $r_v=\sum_{i=1}^k x_{vi}$ is a sufficient statistic. Thus, by conditioning the likelihood onto $\boldsymbol{r}'=(r_1,...,r_n)$, the person parameters $\boldsymbol{\theta}$, which in this context are nuisance parameters, vanish from the likelihood equation, thus, leading to consistently estimated item parameters $\hat{\boldsymbol{\beta}}$.
-
-Some restrictions have to be imposed on the parameters to ensure identifiability. This can be achieved, e.g., by setting certain parameters to zero depending on the model. In the Rasch model one item parameter has to be fixed to 0. This parameter may be considered as baseline difficulty. In addition, in the RSM the category parameters $\omega_0$ and $\omega_1$ are also constrained to 0. In the PCM all parameters representing the first category, i.e. $\beta_{i0}$ with $i=1,\ldots,k$, and [...]
-
-At this point, for the LPCM the likelihood equations with corresponding first and second order derivatives are presented (i.e. \textit{unified CML equations}). In the first version of the \pkg {eRm} package numerical approximations of the Hessian matrix are used. However, to ensure numerical accuracy and to speed up the estimation process, it is planned to implement the analytical solution as given below.
-
-The conditional log-likelihood equation for the LPCM is
-
-\begin{equation}
-\label{eq:cmll}
- \log L_c = \sum_{i=1}^k \sum_{h=1}^{m_i} x_{+ih} \sum_{j=1}^p w_{ihj} \eta_j - \sum_{r=1}^{r_{max}} n_r \log \gamma_r.
-\end{equation}
-
-The maximal raw score is denoted by $r_{max}$ whereas the number of subjects with the same raw score is quoted as $n_r$. Alternatively, by going down to an individual level, the last sum over $r$ can be replaced by $\sum_{v=1}^n \log \gamma_{r_v}$. It is straightforward to show that the LPCM as well as the other extended Rasch models, define an exponential family \citep{And:83}. Thus, the raw score $r_v$ is minimally sufficient for $\theta_v$ and the item totals $x_{.ih}$ are minimally [...]
-
-Crucial expressions are the $\gamma$-terms which are known as \textit{elementary symmetric functions}. More details about these terms are given in the next section. However, in the \pkg {eRm} package the numerically stable \textit{summation algorithm} as suggested by \citet{And:72} is implemented. \citet{FiPo:94} adopted this algorithm for the LPCM and devised also the first order derivative for computing the corresponding derivative of $\log L_c$:
-
-\begin{equation}
-\label{eq:dcml}
-\frac{\partial\log L_c}{\partial\eta_a} = \sum_{i=1}^k \sum_{h=1}^{m_i} w_{iha}\left(x_{+ih} - \epsilon_{ih} \sum_{r=1}^{r_{max}} n_r \frac{ \gamma_{r}^{(i)}}{\gamma_r}\right).
-\end{equation}
-
-It is important to mention that for the CML-representation, the multiplicative Rasch expression is used throughout equations \ref{eq1} to \ref{eq:lpcmeta}, i.e., $\epsilon_i=\exp(-\beta_i)$ for the person parameter. Therefore, $\epsilon_{ih}$ corresponds to the reparameterized item $\times$ category parameter whereas $\epsilon_{ih} > 0$. Furthermore, $\gamma_{r}^{(i)}$ are the first order derivatives of the $\gamma$-functions with respect to item $i$. The index $a$ in $\eta_a$ denotes th [...]
-
-For the second order derivative of $\log L_c$, two cases have to be distinguished: the derivatives for the off-diagonal elements and the derivatives for the main diagonal elements. The item categories with respect to the item index $i$ are coded with $h_i$, and those referring to item $l$ with $h_l$. The second order derivatives of the $\gamma$-functions with respect to items $i$ and $l$ are denoted by $\gamma_r^{(i,l)}$. The corresponding likelihood expressions are
-\begin{align}
-\label{eq:2dcml}
-\frac{\partial\log L_c}{\partial\eta_a \eta_b} = & -\sum_{i=1}^k \sum_{h_i=1}^{m_i} w_{ih_ia}w_{ih_ib}\epsilon_{ih_i} \sum_{r=1}^{r_{max}} n_r \frac{\log \gamma_{r-h_i}}{\gamma_r}\\
-& -\sum_{i=1}^k \sum_{h_i=1}^{m_i} \sum_{l=1}^k \sum_{h_l=1}^{m_l} w_{ih_ia}w_{lh_lb} \left[\epsilon_{ih_i} \epsilon_{lh_l} \left( \sum_{r=1}^{r_{max}} n_r \frac{\gamma_{r}^{(i)}\gamma_{r}^{(l)}}{\gamma_r^2} - \sum_{r=1}^{r_{max}} n_r \frac{\gamma_{r}^{(i,l)}}{\gamma_r}\right)\right]
-\notag
-\end{align}
-for $a\neq b$, and
-\begin{align}
-\label{eq:2dcmlab}
-\frac{\partial\log L_c}{\partial\eta_a^2} = & -\sum_{i=1}^k \sum_{h_i=1}^{m_i} w_{ih_ia}^2 \epsilon_{ih_i} \sum_{r=1}^{r_{max}} n_r \frac{\log \gamma_{r-h_i}}{\gamma_r}\\
-& -\sum_{i=1}^k \sum_{h_i=1}^{m_i} \sum_{l=1}^k \sum_{h_l=1}^{m_l} w_{ih_ia}w_{lh_la}\epsilon_{ih_i} \epsilon_{lh_l}\sum_{r=1}^{r_{max}} n_r \frac{\gamma_{r-h_i}^{(i)}\gamma_{r-h_l}^{(l)}}{\gamma_r^2}
-\notag
-\end{align}
-for $a=b$.
-
-To solve the likelihood equations with respect to $\mathbf{\hat{\eta}}$, a Newton-Raphson algorithm is applied. The update within each iteration step $s$ is performed by
-
-\begin{equation}
-\label{eq:iter}
-\boldsymbol{\hat{\eta}}_s=\boldsymbol{\hat{\eta}}_{s-1}-\mathbf{H}_{s-1}^{-1} \boldsymbol{\delta}_{s-1}.
-\end{equation}
-
-The starting values are $\boldsymbol{\hat{\eta}}_0=\mathbf{0}$.
-$\mathbf{H}_{s-1}^{-1}$ is the inverse of the Hessian matrix composed by
-the elements given in Equation \ref{eq:2dcml} and \ref{eq:2dcmlab} and
-$\boldsymbol{\delta}_{s-1}$ is the gradient at iteration $s-1$ as
-specified in Equation \ref{eq:dcml}. The iteration stops if the
-likelihood difference $\left|\log L_c^{(s)} - \log L_c^{(s-1)}
-\right|\leq \varphi$ where $\varphi$ is a predefined (small) iteration
-limit. Note that in the current version (\Sexpr{packageDescription("eRm", fields = "Version")})
-$\mathbf{H}$ is
-approximated numerically by using the \pkg{nlm} Newton-type algorithm
-provided in the \pkg{stats} package. The analytical solution as given
-in Equation \ref{eq:2dcml} and \ref{eq:2dcmlab} will be implemented in
-the subsequent version of \pkg{eRm}.
-
-
-\subsection{Mathematical properties of the CML estimates}
-\label{sec:mpcml}
-A variety of estimation approaches for IRT models in general and
-for the Rasch model in particular are available: The \emph{joint
-maximum likelihood} (JML) estimation as proposed by \citet{Wright+Panchapakesan:1969}
-which is not recommended since the estimates are not consistent
-\citep[see e.g.][]{Haberman:77}. The basic reason for that is that the
-person parameters $\boldsymbol{\theta}$ are nuisance parameters; the
-larger the sample size, the larger the number of parameters.
-
-A well-known alternative is the \emph{marginal maximum likelihood}
-(MML) estimation \citep{Bock+Aitkin:1981}: A distribution $g(\theta)$ for
-the person parameters is assumed and the resulting situation
-corresponds to a mixed-effects ANOVA: Item difficulties can be
-regarded as fixed effects and person abilities as random effects.
-Thus, IRT models fit into the framework of \emph{generalized linear
-mixed models} (GLMM) as elaborated in \citet{deBoeck+Wilson:2004}. By
-integrating over the ability distribution the random nuisance
-parameters can be removed from the likelihood equations. This leads
-to consistent estimates of the item parameters. Further discussions
-of the MML approach with respect to the CML method will follow.
-
-For the sake of completeness, some other methods for the estimation
-of the item parameters are the following: \citet{CAnd:07} propose
-a Pseudo-ML approach, \citet{Molenaar:1995} and \citet{Linacre:2004} give an
-overview of various (heuristic) non-ML methods, Bayesian
-techniques can be found in \citet[Chapter 7]{BaKi:04}, and for nonparameteric approaches it is referred to \citet{LeVe:86}.
-
-However, back to CML, the main idea behind this approach is the
-assumption that the raw score $r_v$ is a minimal sufficient
-statistic for $\theta_v$. Starting from the equivalent
-multiplicative expression of Equation \ref{eq1} with
-$\xi_v=\exp(\theta_v)$ and $\epsilon_i=\exp(-\beta_i)$, i.e.,
-\begin{equation}
-\label{eq7}
- P(X_{vi}=1)=\frac{\xi_v \epsilon_i}{1+\xi_v \epsilon_i},
-\end{equation}
-the following likelihood for the response pattern $\boldsymbol{x}_v$
-for a certain subject $v$ results:
-\begin{equation}
-\label{eq8}
- P(\boldsymbol{x}_v|\xi_v,\boldsymbol{\epsilon})=\prod_{i=1}^k \frac{(\xi_v \epsilon_i)^{x_{vi}}}{1+\xi_v \epsilon_i}=
- \frac{{\theta_v}^{r_v} \prod_{i=1}^k {\epsilon_i}^{x_{vi}}}{\prod_{i=1}^k (1+\xi_v \epsilon_i)}.
-\end{equation}
-Using the notation $\boldsymbol{y}=(y_1,\ldots ,y_k)$ for all
-possible response patterns with $\sum_{i=1}^k y_i=r_v$, the
-probability for a fixed raw score $r_v$ is
-\begin{equation}
-\label{eq9}
- P(r_v|\xi_v,\boldsymbol{\epsilon})=\sum_{\boldsymbol{y}|r_v} \prod_{i=1}^k \frac{(\xi_v \epsilon_i)^{x_{vi}}}{1+\xi_v \epsilon_i}=\frac{{\theta_v}^{r_v} \sum_{\boldsymbol{y}|r_v} \prod_{i=1}^k {\epsilon_i}^{x_{vi}}}{\prod_{i=1}^k (1+\xi_v \epsilon_i)}.
-\end{equation}
-The crucial term with respect to numerical solutions of the
-likelihood equations is the second term in the numerator:
-\begin{equation}
-\label{eq:gamma}
- \gamma_r(\epsilon_i) \equiv \sum_{\boldsymbol{y}|r_v} \prod_{i=1}^k {\epsilon_i}^{x_{vi}}
-\end{equation}
-These are the \emph{elementary symmetric functions} (of order $r$).
-An overview of efficient computational algorithms and corresponding
-simulation studies can be found in \citet{Li:94}. The \pkg{eRm}
-package uses the summation algorithm as proposed by \citet{And:72}.
-
-Finally, by collecting the different raw scores into the vector
-$\boldsymbol{r}$ the conditional probability of observing response
-pattern $\boldsymbol{x}_v$ with given raw score $r_v$ is
-\begin{equation}
-\label{eq:xraw}
- P(\boldsymbol{x}_v|r_v,\boldsymbol{\epsilon})=\frac{P(\boldsymbol{x}_v|\xi_v,\boldsymbol{\epsilon})}{P(r_v|\xi_v,\boldsymbol{\epsilon})} \,.
-\end{equation}
-By taking the product over the persons (independence assumption),
-the (conditional) likelihood expression for the whole sample becomes
-\begin{equation}
-\label{eq:likall}
- L(\boldsymbol{\epsilon}|\boldsymbol{r})=P(\boldsymbol{x}|\boldsymbol{r},\boldsymbol{\epsilon})=\prod_{v=1}^n \frac{\prod_{i=1}^k {\epsilon_i}^{x_{vi}}}{\gamma_{r_v}}.
-\end{equation}
-With respect to raw score frequencies $n_r$ and by reintroducing the
-$\beta$-parameters, (\ref{eq:likall}) can be reformulated as
-\begin{equation}
-\label{eq12a}
- L(\boldsymbol{\beta}|\boldsymbol{r})= \frac{\exp \left(\sum_{i=1}^k x_{+i}\beta_i \right)}{\prod_{r=0}^k
- \gamma_r^{n_r}} \,,
-\end{equation}
-where $x_{+i}$ are the item raw scores. It is obvious that by
-conditioning the likelihood on the raw scores $\boldsymbol{r}$, the
-person parameters completely vanished from the expression. As a
-consequence, the parameters $\boldsymbol{\hat{\beta}}$ can be
-estimated without knowledge of the subject's abilities. This issue
-is referred as \emph{person-free item assessment} and we will
-discuss this topic within the context of specific objectivity in the
-next section.
-
-Pertaining to asymptotical issues, it can be shown that under mild
-regularity conditions \citep{Pf:94} the CML estimates are
-consistent for $n\rightarrow \infty$ and $k$ fixed, unbiased,
-asymptotically efficient, and normally distributed
-\citep{Andersen:1970}. For the computation of a Rasch model,
-comparatively small samples are sufficient to get reliable estimates
-\citep{Fischer:1988}. Whether the MML estimates are unbiased depends
-on the correct specification of the ability distribution
-$g(\theta)$. In case of an incorrect assumption, the estimates are
-biased which is surely a drawback of this method. If $g(\theta)$ is
-specified appropriately, the CML and MML estimates are
-asymptotically equivalent \citep{Pf:94}.
-
-\citet{Fischer:1981} elaborates on the conditions for the existence and
-the uniqueness of the CML estimates. The crucial condition for the
-data matrix is that $\boldsymbol{X}$ has to be
-\emph{well-conditioned}. To introduce this issue it is convenient to
-look at a matrix which is \emph{ill-conditioned}: A matrix is
-ill-conditioned if there exists a partition of the items into two
-nonempty subsets such that all of a group of subjects responded
-correctly to items $i+1,\ldots,k$ ($\boldsymbol{X}_2$) and all of
-all other subjects failed for items $1,\ldots,i$
-($\boldsymbol{X}_3$), i.e.,
-\begin{table}[h]
-\centering
-\[
-\boldsymbol{X}=
-\left(
-\begin{array}{c|c}
-\boldsymbol{X}_1 & \boldsymbol{X}_2\\
-\hline
-\boldsymbol{X}_3 & \boldsymbol{X}_4\\
-\end{array}
-\right)
-=
-\left(
-\begin{array}{ccc|ccc}
-& & & 1 & \ldots & 1 \\
-& \boldsymbol{X}_1 & & \vdots & \ddots & \vdots \\
-& & & 1 & \ldots & 1 \\
-\hline
-0 & \ldots & 0 & & & \\
-\vdots & \ddots & \vdots & & \boldsymbol{X}_4 & \\
-0 & \ldots & 0 & & & \\
-\end{array}
-\right)
-\]
-\end{table}
-
-Thus, following the definition in \citet{Fischer:1981}: $\boldsymbol{X}$
-will be called \emph{well-conditioned} iff in every possible
-partition of the items into two nonempty subsets some subjects has
-given response 1 on some item in the first set and response 0 on
-some item in the second set. In this case a unique solution for the
-CML estimates $\boldsymbol{\hat{\beta}}$ exists.
-
-This issue is important for structurally incomplete designs which
-often occur in practice; different subsets of items are presented
-to different groups of persons $g=1,\ldots,G$ where $G\leq n$. As a
-consequence, the likelihood values have to be computed for each
-group separately and the joint likelihood is the product over the
-single group likelihoods. Hence, the likelihood in Equation
-\ref{eq12a} becomes
-\begin{equation}
-\label{eq:glik}
-L(\boldsymbol{\beta}|\boldsymbol{r})=\prod_{g=1}^G \frac{\exp \left(\sum_{i=1}^k x_{+i}\beta_i \right)}{\prod_{r=0}^k {\gamma_{g,r}}^{n_{g,r}}}
-\end{equation}
-This also implies the necessity to compute the elementary symmetric
-functions separately for each group. The \pkg{eRm} package can
-handle such structurally incomplete designs.
-
-From the elaborations above it is obvious that from an
-asymptotical point of view the CML estimates are at least as good
-as the MML estimates. In the past, computational problems (speed,
-numerical accuracy) involved in calculating the elementary symmetric
-functions limited the practical usage of the CML approach \citep[see e.g.][]{Gustafsson:1980}.
-Nowadays, these issues are less crucial due to increased computer power.
-
-In some cases MML estimation has advantages not shared by CML: MML
-leads to finite person parameters even for persons with zero and
-perfect raw score, and such persons are not removed from the
-estimation process \citep{Molenaar:1995}. On he other hand the
-consideration of such persons does not seem meaningful from a
-substantial point of view since the person parameters are not
-reliable anymore -- for such subjects the test is too difficult or
-too easy, respectively. Thus, due to these covering effects, a
-corresponding ability estimation is not feasible. However, if the
-research goal is to find ability distributions such persons should
-be regarded and MML can handle this.
-
-When estimates for the person parameters are of interest some care
-has to be taken if the CML method is used since person parameters
-cancel from the estimation equations. Usually, they are estimated
-(once having obtained values for the item parameters) by inserting
-$\boldsymbol{\hat{\beta}}$ (or equivalently
-$\boldsymbol{\hat{\epsilon}}$) into Equation \ref {eq8} and
-solving with respect to $\boldsymbol{\theta}$. Alternatively,
-Bayesian procedures are applicable \citep{Hoijtink+Boomsma:1995}. It is again
-pointed out that each person in the sample gets an own parameter
-even though limited by the number of different raw scores.
-
-\subsection{CML and specific objectivity}
-In general, the Rasch model can be regarded as a measurement model:
-Starting from the (nominally scaled) 0/1-data matrix
-$\boldsymbol{X}$, the person raw scores $r_v$ are on an ordinal
-level. They, in turn, are used to estimate the item parameters
-$\boldsymbol{\beta}$ which are on an interval scale provided that
-the Rasch model holds.
-
-Thus, Rasch models allow for comparisons between objects on an
-interval level. Rasch reasoned on requirements to be fulfilled such
-that a specific proposition within this context can be regarded as
-``scientific''. His conclusions were that a basic requirement is the
-``objectivity'' of comparisons \citep{Ra:61}. This claim
-contrasts assumptions met in \emph{classical test theory} (CTT). A
-major advantage of the Rasch model over CTT models is the
-\emph{sample independence} of the results. The relevant concepts in
-CTT are based on a linear model for the ``true score" leading to
-some indices, often correlation coefficients, which in turn depend
-on the observed data. This is a major drawback in CTT. According to
-\citet{Fisch:74}, sample independence in IRT models has the
-following implications:
-\begin{itemize}
- \item The person-specific results (i.e., essentially $\boldsymbol{\theta}$) do not depend on the assignment of a person to a certain subject group nor on the selected test items from an item pool $\Psi$.
- \item Changes in the skills of a person on the latent trait can be determined independently from its base level and independently from the selected item subset $\psi \subset \Psi$.
- \item From both theoretical and practical perspective the requirement for representativeness of the sample is obsolete in terms of a true random selection process.
-\end{itemize}
-Based on these requirements for parameter comparisons, \citet{Ra:77}
-introduced the term \emph{specific objectivity}: \emph{objective}
-because any comparison of a pair of parameters is independent of any
-other parameters or comparisons; \emph{specifically objective}
-because the comparison made was relative to some specified frame of
-reference \citep{Andrich:88}. In other words, if specific
-objectivity holds, two persons $v$ and $w$ with corresponding
-parameters $\theta_v$ and $\theta_w$, are comparable independently
-from the remaining persons in the sample and independently from the
-presented item subset $\psi$. In turn, for two items $i$ and $j$
-with parameters $\beta_i$ and $\beta_j$, the comparison of these
-items can be accomplished independently from the remaining items in
-$\Psi$ and independently from the persons in the sample.
-
-The latter is crucial since it reflects completely what is called
-sample independence. If we think not only of comparing $\beta_i$ and
-$\beta_j$ but rather to estimate these parameters, we achieve a
-point where specific objectivity requires a procedure which is able
-to provide estimates $\boldsymbol{\hat{\beta}}$ that do not
-depend on the sample. This implies that
-$\boldsymbol{\hat{\beta}}$ should be computable without the
-involvement of $\boldsymbol{\theta}$. CML estimation fulfills this requirement: By
-conditioning on the sufficient raw score vector $\boldsymbol{r}$,
-$\boldsymbol{\theta}$ disappears from the likelihood equation and
-$L(\boldsymbol{\beta}|\boldsymbol{r})$ can be solved without
-knowledge of $\boldsymbol{\theta}$. This issue is referred to as
-\emph{separability of item and person parameters} \citep[see e.g.][]{Wright+Masters:1982}. Furthermore, separability implies that no specific distribution should be assumed neither for the person nor for the item parameters \citep{Rost:2000}. MML estimation requires such assumptions. At this point it is clear that CML estimation is
-the only estimation method within the Rasch measurement context
-fulfilling the requirement of \emph{person-free item calibration}
-and, thus, it maps the epistemological theory of specific
-objectivity to a statistical maximum likelihood framework. Note that
-strictly speaking any statistical result based on sample
-observations is sample-dependent because any result depends at least
-on the sample size \citep{Fischer:1987}. The estimation of the item
-parameters is ``sample-independent", a term indicating the fact that
-the actually obtained sample of a certain population is not of
-relevance for the statistical inference on these parameters
-\citep[][p. 23]{Kubinger:1989}.
-
-\subsection{Estimation of person parameters}
-CML estimation for person parameters is not recommended due to computational issues. The \pkg{eRm} package provides two methods for this estimation. The first is ordinary ML where the CML-based item parameters are plugged into the joint ML equation. The likelihood is optimized with respect to $\boldsymbol{\theta}$. \citet{And:95} gives a general formulation of this ML estimate with $r_v=r$ and $\theta_v=\theta$:
-\begin{equation}
-\label{eq17}
- r - \sum_{i=1}^k \sum_{h=1}^{m_i} \frac{h \exp(h \theta+\hat{\beta}_{ih})}{\sum_{l=0}^{m_i}\exp(h \theta_v+\hat{\beta}_{il})}=0
-\end{equation}
-
-\citet{Warm:1989} proposed a weighted likelihood estimation (WLE) which is more accurate compared to ML. For the dichotomous Rasch model the expression to be solved with respect to $\boldsymbol{\theta}$ is
-\begin{equation}
-P(\theta_v|\boldsymbol{x}_v, \hat{\boldsymbol{\beta}}) \propto \frac{exp(r_v\theta_v)}{\prod_i (1+exp(\theta_v-\hat{\beta}_i)}\sum_i p_{vi}(1-p_{vi})
-\end{equation}
-Again, the item parameter vector $\hat{\boldsymbol{\beta}}$ is used from CML. This approach will implemented in a subsequent \pkg{eRm} version. Additional explanations and simulation studies regarding person parameter estimation can be found in \citet{Hoijtink+Boomsma:1995}.
-
-%----------------- end parameter estimation -----------------
-
-\section{Testing extended Rasch models}
-\label{Gof}
-
-Testing IRT models involves two parts: First, item- and person-wise
-statistics can be examined; in particular item-fit and person-fit
-statistics. Secondly, based on CML properties, various model tests
-can be derived \citep[see][]{Glas+Verhelst:1995a,
-Glas+Verhelst:1995b}.
-
-\subsection{Item-fit and person-fit statistics}
-
-Commonly in IRT, items and persons are excluded due to item-fit and
-person-fit statistics. Both are residual based measures: The
-observed data matrix $\mathbf{X}$ is compared with the model
-probability matrix $\mathbf{P}$. Computing standardized residuals
-for all observations gives the $n \times k$ residual matrix
-$\mathbf{R}$. The squared column sums correspond to item-fit
-statistics and the squared row sums to person-fit statistics both of
-which are $\chi^2$-distributed with the corresponding degrees of
-freedom. Based on these quantities unweighted (\textsl{outfit}) and
-weighted (\textsl{infit}) mean-square statistics can also be used to
-evaluate item and person fit \citep[see
-e.g.][]{Wright+Masters:1982}.
-
-\subsection{A Wald test for item elimination}
-A helpful implication of CML estimates is that subsequent test
-statistics are readily obtained and model tests are easy to carry
-out. Basically, we have to distinguish between test on item level
-and global model tests.
-
-On item level, sample independence reflects the property that by
-splitting up the sample in, e.g., two parts, the corresponding
-parameter vectors $\boldsymbol{\hat{\beta}}^{(1)}$ and
-$\boldsymbol{\hat{\beta}}^{(2)}$ should be the same. Thus, when
-we want to achieve Rasch model fit those items have to be
-eliminated from the test which differ in the subsamples. This
-important issue in test calibration can be examined, e.g., by using
-a graphical model test. \citet{FiSch:70} propose a $N(0,1)$-distributed
-test statistic which compares the item parameters for two subgroups:
-\begin{equation}
-\label{eq:wald}
- z=\frac{\beta_i^{(1)}-\beta_i^{(2)}}{\sqrt{Var_i^{(1)}-Var_i^{(2)}}}
-\end{equation}
-The variance term in the denominator is based on Fisher's function of ``information in the sample".
-However, as \citet{Glas+Verhelst:1995a} point out
-discussing their Wald-type test that this term can be extracted directly
-from the variance-covariance matrix of the CML estimates. This Wald approach is provided in \pkg{eRm} by means of the function \code{Waldtest()}.
-
-\subsection{Andersen's likelihood-ratio test}
-In the \pkg {eRm} package the likelihood ratio test statistic $LR$, initially proposed by \citet{And:73} is computed for the RM, the RSM, and the PCM. For the models with linear extensions, $LR$ has to be computed separately for each measurement point and subgroup.
-\begin{equation}
-\label{eq15}
-LR = 2\left(\sum_{g=1}^G \log L_c(\boldsymbol{\hat{\eta}}_g;\boldsymbol{X}_g)-\log L_c(\boldsymbol{\hat{\eta}};\boldsymbol{X})\right)
-\end{equation}
-The underlying principle of this test statistic is that of \textit{subgroup homogeneity} in Rasch models: for arbitrary disjoint subgroups $g=1,...,G$ the parameter estimates $\boldsymbol{\hat{\eta}}_g$ have to be the same. $LR$ is asymptotically $\chi^2$-distributed with $df$ equal to the number of parameters estimated in the subgroups minus the number of parameters in the total data set. For the sake of computational efficiency, the \pkg {eRm} package performs a person raw score median [...]
-
-\subsection{Nonparametric (``exact'') Tests}
-Based on the package \pkg{RaschSampler} by
-\citet{Verhelst+Hatzinger+Mair:2007} several Rasch model tests as
-proposed by \citep{Ponocny:2001} are provided.
-
-\subsection{Martin-L\"of Test}
-Applying the LR principle to subsets of items, Martin-L\"of \citep[1973, see][]{Glas+Verhelst:1995a} suggested a statistic to
-evaluate if two groups of items are homogeneous, i.e.,
-to test the unidimensionality axiom.
-%-------------------------- end goodness-of-fit ------------------
-
-%---------------------------- APPLIED SECTION ----------------------------
-\section{The eRm package and application examples}
-\label{sec:pack}
-The underlying idea of the \pkg {eRm} package is to provide a user-friendly
-flexible tool to compute extended Rasch models. This implies, amongst others,
-an automatic generation of the design matrix $\mathbf{W}$. However, in order to
-test specific hypotheses the user may specify $\mathbf{W}$ allowing the package
-to be flexible enough for computing IRT-models beyond their regular applications.
-In the following subsections, various examples are provided pertaining to different model and design
-matrix scenarios. Due to intelligibility matters, the artificial data sets are kept rather small. A detailed description in German of applications of various extendend Rasch models using the \pkg{eRm} package can be found in \citet{Poinstingl+Mair+Hatzinger:07}.
-
-\subsection{Structure of the eRm package}
-Embedding \pkg{eRm} into the flexible framework of \proglang{R} is a
-crucial benefit over existing stand-alone programs like WINMIRA
-\citep{Davier:1998}, LPCM-WIN \citep{FiPS:98}, and others.
-
-Another important issue in the development phase was that the
-package should be flexible enough to allow for CML compatible
-polytomous generalizations of the basic Rasch model such as the RSM
-and the PCM. In addition, by introducing a design matrix concept
-linear extensions of these basic models should be applicable. This
-approach resulted in including the LLTM, the LRSM and the LPCM as
-the most general model into the \pkg{eRm} package. For the latter
-model the CML estimation was implemented which can be used for the
-remaining models as well. A corresponding
-graphical representation is given in Figure \ref{fig:body}.
-
-\begin{figure}[hbt]
-\begin{center}
- \includegraphics[width=13.7cm, height=6.5cm]{UCML.jpg}
- \caption{\label{fig:body}Bodywork of the \pkg{eRm} routine}
-\end{center}
-\end{figure}
-
-An important benefit of the package with respect to linearly
-extended models is that for certain models the design matrix
-$\boldsymbol{W}$ can be generated automatically \citep[LPCM-WIN,][]{FiPS:98} also allows for specifying design matrices but in
-case of more complex models this can become a tedious task and the
-user must have a thorough understanding of establishing proper
-design structures). For repeated measurement models time contrasts
-in the \pkg{eRm} can be simply specified by defining the number of
-measurement points, i.e., {\tt mpoints}. To regard group contrasts
-like, e.g., treatment and control groups, a corresponding vector
-({\tt groupvec}) can be specified that denotes which person belongs
-to which group. However, $\boldsymbol{W}$ can also be defined by the
-user.
-
-A recently added feature of the routine is the option to allow for
-structurally missing values. This is required, e.g., in situations
-when different subsets of items are presented to different groups of
-subjects as described in Section \ref{sec:mpcml}. These person groups
-are identified automatically: In the data matrix $\boldsymbol{X}$,
-those items which are not presented to a certain subject are
-declared as \code{NA}s, as usual in \proglang{R}.
-
-After solving the CML equations by the Newton-Raphson method, the
-output of the routine consists of the ``basic" parameter estimates
-$\boldsymbol{\hat{\eta}}$, the corresponding variance-covariance
-matrix, and consequently the vector with the standard errors.
-Furthermore, the ordinary item parameter estimates
-$\boldsymbol{\hat{\beta}}$ are computed by using the linear
-transformation
-$\boldsymbol{\hat{\beta}}=\boldsymbol{W}\boldsymbol{\hat{\eta}}$.
-For ordinary Rasch models these basic parameters correspond to the
-item easiness. For the RM, the RSM, and the PCM, however, we display
-$\boldsymbol{\hat{\eta}}$ as $\boldsymbol{-\hat{\eta}}$, i.e., as difficulty.
-It has to be mentioned that the CML equation is
-solved with the restriction that one item parameter has to be fixed
-to zero (we use
- $\beta_1=0$). For the sake of interpretability, the resulting
-estimates $\boldsymbol{\hat{\beta}}$ can easily be transformed
-into ``sum-zero" restricted $\boldsymbol{\hat{\beta}^*}$ by
-applying
-$\hat{\beta}_i^*=\hat{\beta}_i-\sum_i{\hat{\beta}_i}/k$.
-This transformation is also used for the graphical model test.
-
-\subsection{Example 1: Rasch model}
-We start the example section
-with a simple Rasch model based on a $100 \times 30$ data matrix.
-First, we estimate the item parameters using the function
-\code{RM()} and then the person parameters with
-\code{person.parameters()}.
-
-<<>>=
-library(eRm)
-data(raschdat1)
-res.rasch <- RM(raschdat1)
-pres.rasch <- person.parameter(res.rasch)
-@
-
-Then we use Andersen's LR-test for goodness-of-fit with mean split criterion:
-<<>>=
-lrres.rasch <- LRtest(res.rasch, splitcr = "mean", se = TRUE)
-lrres.rasch
-@
-
-We see that the model fits and a graphical representation of this
-result (subset of items only) is given in Figure \ref{fig:GOF} by means
-of a goodness-of-fit plot with confidence ellipses.
-
-\begin{figure}[hbt]
-\begin{center}
-<<fig = TRUE>>=
-plotGOF(lrres.rasch, beta.subset=c(14,5,18,7,1), tlab="item", conf=list(ia=FALSE,col="blue",lty="dotted"))
-@
-\caption{\label{fig:GOF} Goodness-of-fit plot for some items with confidence ellipses.}
-\end{center}
-\end{figure}
-
-To be able to draw confidence ellipses it is needed to set \code{se = TRUE} when computing the LR-test.
-
-\subsection{Example 2: LLTM as a restricted Rasch model}
-As mentioned in Section \ref{Rep}, also the models with the linear extensions on
-the item parameters can be seen as special cases of their underlying basic model.
-In fact, the LLTM as presented below and following the original idea by \citet{Scheib:72},
-is a restricted RM, i.e. the number of estimated parameters is smaller compared to a Rasch model. The data matrix
-$\mathbf{X}$ consists of $n=15$ persons and $k=5$ items. Furthermore, we specify a design matrix $\mathbf{W}$ (following Equation \ref{eq4}) with specific weight elements $w_{ij}$.
-
-<<>>=
-data(lltmdat2)
-W <- matrix(c(1,2,1,3,2,2,2,1,1,1),ncol=2)
-res.lltm <- LLTM(lltmdat2, W)
-summary(res.lltm)
-@
-
-The \code{summary()} method provides point estimates and standard
-errors for the basic parameters and for the resulting item
-parameters. Note that item parameters in \pkg{eRm} are always
-estimated as easiness parameters according to equations \ref{eq1}
-and \ref{eq2} but not \ref{eq:rasch}. If the sign is switched, the
-user gets difficulty parameters (the standard errors remain the
-same, of course). However,
-all plotting functions \code{plotGOF}, \code{plotICC},
-\code{plotjointICC}, and \code{plotPImap}, as well as the function
-\code{thresholds} display the difficulty parameters. The same applies
-for the basic parameters $\eta$ in the output of the RM, RSM, and PCM.
-
-\subsection{Example 3: RSM and PCM}
-Again, we provide an artificial data set now with $n=300$ persons and $k=4$ items;
-each of them with $m+1=3$ categories. We start with the estimation of an RSM and, subsequently,
-we calculate the corresponding category-intersection parameters using the function \code{thresholds()}.
-
-<<>>=
-data(pcmdat2)
-res.rsm <- RSM(pcmdat2)
-thresholds(res.rsm)
-@
-
-The location parameter is basically the item difficulty and the thesholds are the points in the
-ICC plot given in Figure \ref{fig:ICC} where the category curves intersect:
-
-<<fig = FALSE>>=
-plotICC(res.rsm, mplot=TRUE, legpos=FALSE,ask=FALSE)
-@
-
-\begin{figure}[hbt]
-\begin{center}
-<<fig = TRUE, echo=FALSE>>=
-plotICC(res.rsm, mplot=TRUE, legpos=FALSE,ask=FALSE)
-@
-\caption{\label{fig:ICC} ICC plot for an RSM.}
-\end{center}
-\end{figure}
-
-The RSM restricts the threshold distances to be the same across all items.
-This strong assumption can be relaxed using a PCM. The results are represented in a person-item map
-(see Figure \ref{fig:PImap}).
-
-<<fig=FALSE>>=
-res.pcm <- PCM(pcmdat2)
-plotPImap(res.pcm, sorted = TRUE)
-@
-
-\begin{figure}[hbt]
-\begin{center}
-<<fig=TRUE,echo=FALSE>>=
-res.pcm <- PCM(pcmdat2)
-plotPImap(res.pcm, sorted = TRUE)
-@
-\caption{\label{fig:PImap} Person-Item map for a PCM.}
-\end{center}
-\end{figure}
-
-After estimating the person parameters we can check the item-fit statistics.
-<<>>=
-pres.pcm <- person.parameter(res.pcm)
-itemfit(pres.pcm)
-@
-
-A likelihood ratio test comparing the RSM and the PCM indicates that the PCM provides a better fit.
-%Since none of the items is significant we can conclude that the data fit the PCM.
-
-<<>>=
-lr<- 2*(res.pcm$loglik-res.rsm$loglik)
-df<- res.pcm$npar-res.rsm$npar
-pvalue<-1-pchisq(lr,df)
-cat("LR statistic: ", lr, " df =",df, " p =",pvalue, "\n")
-@
-
-
-\subsection{An LPCM for repeated measurements in different groups}
-The most complex example refers to an LPCM with two measurement points.
-In addition, the hypothesis is of interest whether the treatment has an effect.
-The corresponding contrast is the last column in $\mathbf{W}$ below.
-
-First, the data matrix $\mathbf{X}$ is specified. We assume an artificial test consisting of $k=3$ items
-which was presented twice to the subjects. The first 3 columns in $\mathbf{X}$ correspond
-to the first test occasion, whereas the last 3 to the second occasion.
-Generally, the first $k$ columns correspond to the first test occasion, the next $k$ columns for the second, etc.
-In total, there are $n=20$ subjects. Among these, the first 10 persons belong to the first group (e.g., control),
-and the next 10 persons to the second group (e.g., treatment). This is specified
-by a group vector:
-
-<<>>=
-data(lpcmdat)
-grouplpcm <- rep(1:2, each = 10)
-@
-
-Again, $\boldsymbol{W}$ is generated automatically. In general, for such designs
-the generation of $\boldsymbol{W}$ consists first of the item contrasts,
-followed by the time contrasts and finally by the group main effects except for
-the first measurement point (due to identifiability issues, as already described).
-
-<<>>=
-reslpcm <- LPCM(lpcmdat, mpoints = 2, groupvec = grouplpcm, sum0 = FALSE)
-model.matrix(reslpcm)
-@
-
-The parameter estimates are the following:
-
-<<echo = FALSE>>=
-reslpcm
-@
-
-Testing whether the $\eta$-parameters equal 0 is mostly not of relevance for those
-parameters referring to the items (in this example $\eta_1,...,\eta_8$).
-But for the remaining contrasts, $H_0: \eta_9=0$ (implying no general time effect)
-can not be rejected ($p=.44$), whereas hypothesis $H_0: \eta_{10}=0$ has to be rejected
-($p=.004$) when applying a $z$-test.
-This suggests that there is a significant treatment effect over the measurement points.
-If a user wants to perform additional tests such as a Wald test for the equivalence
-of two $\eta$-parameters, the \code{vcov} method can be applied to get the
-variance-covariance matrix.
-
-\section{Additional topics}
-
-This section will be extended successively with new developments and
-components which do not directly relate to the modeling core of
-\pkg{eRm} but may prove to be useful add-ons.
-
-\subsection{The eRm simulation module}
-A recent \pkg{eRm} development is the implementation of a simulation module to generate 0-1 matrices for different Rasch scenarios. In this article we give a brief overview about the functionality and for more detailed descriptions (within the context of model testing) it is referred to \citet{Mair:2006} and \citet{Suarez+Glas:2003}.
-
-For each scenario the user has the option either to assign $\boldsymbol{\theta}$ and $\boldsymbol{\beta}$ as vectors to the simulation function (e.g. by drawing parameters from a uniform distribution) or to let the function draw the parameters from a $N(0,1)$ distribution. The first scenario is the simulation of Rasch homogenous data by means of the function \code{sim.rasch()}. The parameter values are plugged into equation \ref{eq:rasch} and it results the matrix $\mathbf{P}$ of model p [...]
-\begin{equation*}
-x_{vi} = \left\{
- \begin{array}{rl}
- 1 & \text{if } p^{\star}_{vi} \leq p_{vi}\\
- 0 & \text{if } p^{\star}_{vi} > p_{vi}\\
- \end{array} \right.
-\end{equation*}
-Alternatively, the user can specify a fixed cutpoint $p^{\star}:=p^{\star}_{vi}$ (e.g. $p^{\star} = 0.5$) and make the decision according to the same rule. This option is provided by means of the \code{cutpoint} argument. Caution is advised when using this deterministic option since this leads likely to ill-conditioned data matrices.
-
-The second scenario in this module regards the violation of the parallel ICC assumption which leads to the two-parameter logistic model (2-PL) proposed by \citet{Birnbaum:1968}:
-\begin{equation}
-\label{eq:2pl}
- P(X_{vi}=1)=\frac{\exp(\alpha_i(\theta_v - \beta_i))}{1+\exp(\alpha_i(\theta_v-\beta_i))}.
-\end{equation}
-The parameter $\alpha_i$ denotes the item discrimination which for the Rasch model is 1 across all items. Thus, each item score gets a weight and the raw scores are not sufficient anymore. The function for simulating 2-PL data is \code{sim.2pl()} and if $\boldsymbol{\alpha}$ is not specified by the user by means of the argument \code{discrim}, the discrimination parameters are drawn from a log-normal distribution. The reasons for using this particular kind of distribution are the followi [...]
-ns of the dispersion parameter $\sigma^2$. A value of $\sigma^2 = .50$ already denotes a strong violation. The lower $\sigma^2$, the closer the values lie around 1. In this case the $\alpha_i$ are close to the Rasch slopes.
-
-Using the function \code{sim.xdim()} the unidimensionality assumptions is violated. This function allows for the simulation of multidimensional Rasch models as for instance given \citet{Glas:1992} and \citet{Adams+Wilson+Wang:1997}. Multidimensionality implies that one single item measures more than one latent construct. Let us denote the number of these latent traits by $D$. Consequently, each person has a vector of ability parameters $\boldsymbol{\theta}_v$ of length $D$. These vectors [...]
- is not provided by the user, \code{sim.xdim()} generates $\mathbf{Z}$ such that each $\mathbf{z}_i$ contains only nonzero element which indicates the assigned dimension. This corresponds to the \emph{between-item multidimensional model} \citep{Adams+Wilson+Wang:1997}. However, in any case the person part of the model is $\mathbf{z}_i^T \boldsymbol{\theta}_v$ which replaces $\theta_v$ in Equation \ref{eq:rasch}.
-
-Finally, locally dependent item responses can be produced by means of the function \code{sim.locdep()}. Local dependence implies the introduction of pair-wise item correlations $\delta_{ij}$. If these correlations are constant across items, the argument \code{it.cor} can be a single value $\delta$. A value $\delta = 0$ corresponds to the Rasch model whereas $\delta = 1$ leads to the strongest violation. Alternatively, for different pair-wise item correlations, the user can specify a VC-m [...]
-\begin{equation}
-P(X_{vi}=1|X_{vj}=x_{vj})=\frac{\exp(\theta_v - \beta_i + x_{vj}\delta_{ij})}{1+\exp(\theta_v-\beta_i + x_{vj}\delta_{ij})}.
-\end{equation}
-This model was proposed by \citet{Jannarone:1986} and is suited to model locally dependent item responses.
-
-
-\section{Discussion and outlook}
-\label{sec:disc}
-
-Here we give a brief outline of future \pkg{eRm} developments. The
-CML estimation approach, in combination with the EM-algorithm, can
-also be used to estimate \textit{mixed Rasch models} (MIRA). The
-basic idea behind such models is that the extended Rasch model holds
-within subpopulations of individuals, but with different parameter
-values for each subgroup. Corresponding elaborations are given in
-\citet{RoDa:95}.
-
-In Rasch models the item discrimination parameter $\alpha_i$ is
-always fixed to 1 and thus it does not appear in the basic
-equation. Allowing for different discrimination parameters across
-items leads to the two-parameter logistic model as given in Equation
-\ref{eq:2pl}. In this model the raw scores are not sufficient
-statistics anymore and hence CML can not be applied. 2-PL models can
-be estimated by means of the \pkg{ltm} package \citep{Riz:06}.
-However, \citet{Verhelst+Glas:1995} formulated the one parameter
-logistic model (OPLM) where the $\alpha_i$ do not vary across the
-items but are unequal to one. The basic strategy to estimate OPLM is
-a three-step approach: First, the item parameters of the Rasch model
-are computed. Then, discrimination parameters are computed under
-certain restrictions. Finally, using these discrimination weights,
-the item parameters for the OPLM are estimated using CML. This is a
-more flexible version of the Rasch model in terms of different
-slopes.
-
-To conclude, the \pkg{eRm} package is a tool to estimate extended
-Rasch models for unidimensional traits. The generalizations towards
-different numbers of item categories, linear extensions to allow for
-introducing item covariates and/or trend and optionally group
-contrasts are important issues when examining item behavior and
-person performances in tests. This improves the feasibility of IRT
-models with respect to a wide variety of application areas.
-
-\bibliography{eRmvig}
-
-\end{document}
diff --git a/inst/doc/eRmvig.bib b/inst/doc/eRmvig.bib
old mode 100644
new mode 100755
diff --git a/inst/doc/eRmvig.pdf b/inst/doc/eRmvig.pdf
deleted file mode 100644
index d9d1b73..0000000
Binary files a/inst/doc/eRmvig.pdf and /dev/null differ
diff --git a/inst/doc/index.html b/inst/doc/index.html.old
old mode 100644
new mode 100755
similarity index 100%
rename from inst/doc/index.html
rename to inst/doc/index.html.old
diff --git a/inst/doc/jss.bst b/inst/doc/jss.bst
old mode 100644
new mode 100755
diff --git a/inst/doc/modelhierarchy.pdf b/inst/doc/modelhierarchy.pdf
old mode 100644
new mode 100755
diff --git a/man/IC.Rd b/man/IC.Rd
old mode 100644
new mode 100755
diff --git a/man/LLRA.Rd b/man/LLRA.Rd
new file mode 100755
index 0000000..64d7426
--- /dev/null
+++ b/man/LLRA.Rd
@@ -0,0 +1,136 @@
+\name{LLRA}
+\alias{LLRA}
+\alias{print.llra}
+
+\title{Fit Linear Logistic Models with Relaxed Assumptions (LLRA)
+}
+\description{
+Automatically builds design matrix and fits LLRA.
+}
+\usage{
+LLRA(X, W, mpoints, groups, baseline, itmgrps = NULL, ...)
+
+\method{print}{llra}(x, ...)
+}
+\arguments{
+ \item{X}{Data matrix as described in Hatzinger and Rusch (2009). It
+ must be of wide format, e.g. for each person all item answers are
+ written in columns for t1, t2, t3 etc. Hence each row corresponds to
+ all observations for a single person. See llraDat1 for an example.
+ Missing values are not allowed.
+ }
+ \item{W}{Design Matrix for LLRA to be passed to \code{LPCM}. If missing, it
+ is generated automatically.
+ }
+ \item{mpoints}{The number of time points.
+ }
+ \item{groups}{Vector, matrix or data frame with subject/treatment
+ covariates.
+ }
+ \item{baseline}{An optional vector with the baseline values for the
+ columns in group.
+ }
+ \item{itmgrps}{
+Specifies how many groups of items there are. Currently not functional but may be useful in the future.
+}
+ \item{x}{For the print method, an object of class \code{"llra"}.
+}
+ \item{\dots}{
+Additional arguments to be passed to and from other methods.
+}
+}
+\details{The function \code{LLRA} is a wrapper for \code{LPCM} to fit
+ Linear Logistic Models with Relaxed Assumptions (LLRA). LLRA
+ are extensions of the LPCM for the measurement of change over a number
+ of discrete time points for a set of
+ items. It can incorporate categorical covariate information. If no
+ design matrix W is passed as an argument, it is built automatically
+ from scratch.
+
+ Unless passed by the user, the baseline group is always the one with
+ the lowest (alpha-)numerical value for argument \code{groups}. All
+ other groups are labeled decreasingly according to the
+ (alpha)-numerical value, e.g. with 2 treatment groups (TG1 and TG2)
+ and one control group (CG), CG will be the baseline than TG1 and TG2.
+ Hence the group effects are ordered like
+ \code{rev((unique(names(groupvec)))} for naming.
+
+ Caution is advised as LLRA will fail if all changes for a group will be into a
+ single direction (e.g. all subjects in the treatment group show
+ improvement). Currently only data matrices are supported as arguments.
+}
+\value{
+Returns an object of class \code{"llra"} (also inheriting from class
+\code{"eRm"}) containing
+
+\item{loglik}{Conditional log-likelihood.}
+\item{iter}{Number of iterations.}
+\item{npar}{Number of parameters.}
+\item{convergence}{See code output in nlm.}
+\item{etapar}{Estimated basic item parameters. These are the LLRA
+effect parameters.}
+\item{se.eta}{Standard errors of the estimated basic item parameters.}
+\item{betapar}{Estimated item (easiness) parameters of the virtual
+items (not useful for interpretation here).}
+\item{se.beta}{Standard errors of virtual item parameters (not useful for interpretation here).}
+\item{hessian}{Hessian matrix if \code{se = TRUE}.}
+\item{W}{Design matrix.}
+\item{X}{Data matrix in long format. The columns correspond to the
+ measurement points and each persons item answers are listed
+ susequently in rows.}
+\item{X01}{Dichotomized data matrix.}
+\item{groupvec}{Assignment vector.}
+\item{call}{The matched call.}
+\item{itms}{The number of items.}
+}
+\references{
+Fischer, G.H. (1995) Linear logistic models for change. In G.H. Fischer
+and I. W. Molenaar (eds.), \emph{Rasch models: Foundations, recent
+developments and applications} (pp. 157--181), New York: Springer.
+
+Glueck, J. and Spiel, C. (1997) Item response models for repeated
+measures designs: Application and limitations of four different
+approaches. \emph{Methods of Psychological
+ Research}, \bold{2}. \url{http://www.dgps.de/fachgruppen/methoden/mpr-online/issue2/art6/article.html}
+
+Hatzinger, R. and Rusch, T. (2009) IRT models with relaxed assumptions
+in eRm: A manual-like instruction. \emph{Psychology Science Quarterly}, \bold{51},
+pp. 87--120, \url{http://erm.r-forge.r-project.org/psq_1_2009_06_87-120.pdf}
+}
+\author{
+Thomas Rusch
+}
+\section{Warning}{A warning is printed that the first two categories
+ for polytomous items are equated to save parameters. See Hatzinger and
+ Rusch (2009) for a justification why this is valid also from a substantive
+ point of view.}
+\seealso{
+The function to build the design matrix \code{\link{build_W}}, and the
+S3 methods \code{\link{summary.llra}} and \code{\link{plotTR}} and
+\code{\link{plotGR}} for plotting.
+}
+\examples{
+ ##Example 6 from Hatzinger & Rusch (2009)
+ data("llradat3")
+ groups <- c(rep("TG",30),rep("CG",30))
+ llra1 <- LLRA(llradat3,mpoints=2,groups=groups)
+ llra1
+
+ ##An LLRA with 2 treatment groups and 1 baseline group, 5 items and 4
+ ##time points. Item 1 is dichotomous, all others have 3, 4, 5, 6
+ ##categories respectively.
+\dontrun{
+ data("llraDat2")
+ dats <- llraDat2[1:20]
+ groups <- llraDat2$group
+ tps <- 4
+ ex2 <- LLRA(dats,mpoints=tps,groups=groups) #baseline CG
+ #baseline TG1
+ ex2a <- LLRA(dats,mpoints=tps,groups=groups,baseline="TG1") #baseline TG1
+ ex2
+ summary(ex2)
+ summary(ex2a)
+ plotGR(ex2)
+ plotTR(ex2)
+}
+}
diff --git a/man/LLTM.Rd b/man/LLTM.Rd
old mode 100644
new mode 100755
diff --git a/man/LPCM.Rd b/man/LPCM.Rd
old mode 100644
new mode 100755
diff --git a/man/LRSM.Rd b/man/LRSM.Rd
old mode 100644
new mode 100755
diff --git a/man/LRtest.Rd b/man/LRtest.Rd
old mode 100644
new mode 100755
index daeb790..a7437b9
--- a/man/LRtest.Rd
+++ b/man/LRtest.Rd
@@ -47,13 +47,15 @@
\item{conf}{for plotting confidence ellipses for the item parameters. If \code{conf=NULL}
(the default) no ellipses are drawn. Otherwise, \code{conf} must be
specified as a list with optional elements: \code{gamma}, is
- the confidence level (numeric), \code{col} and \code{lty}, colour and linetype (see \code{\link{par}}),
+ the confidence level (numeric), \code{col} and \code{lty},
+ colour and linetype (see \code{\link{par}}), \code{which} (numeric index vector) specifying for which
+ items ellipses are drawn (must be a subset of \code{beta.subset}),
and \code{ia}, logical, if the ellipses are to be drawn interactively (cf.
\code{tlab="identify"} above). If \code{conf} is specified as a an empty list, %\code{conf=list()},
the default values \code{conf=list(gamma=0.95, col="red", lty="dashed", ia=FALSE)}
will be used. See example below. To use \code{conf}, the LR object \code{x} has
to be generated using the option \code{se=TRUE} in \code{LRtest()}.
- }
+ For specification of \code{col} and \code{which} see Details and Examples below.}
\item{ctrline}{for plotting confidence bands (control lines, cf.eg.Wright and Stone, 1999).
If \code{ctrline=NULL}
(the default) no lines are drawn. Otherwise, \code{ctrline} must be
@@ -76,6 +78,16 @@
by clicking the second (right) mouse button and selecting `Stop' from the menu, or from the `Stop'
menu on the graphics window.
+ Using the specification \code{which} in allows for selectively drawing ellipses for
+ certain items only, e.g., \code{which=1:3} draws ellipses for items 1 to 3 (as long as they are included
+ in \code{beta.subset}). The default is drawing ellipses for all items.
+ The element \code{col} in the \code{conf} list can either be a single colour
+ specification such as \code{"blue"} or a vector with colour specifications for all items.
+ The length must be the same as the number of ellipses to be drawn. For colour specification
+ a palette can be set up using standard palettes (e.g. \code{\link{rainbow}}) or palettes from
+ the \code{colorspace} or \code{RColorBrewer} package. An example is given below.
+
+
\code{summary} and \code{print} methods are available for objects of class \code{LR}.
}
\value{
@@ -132,6 +144,15 @@ plotGOF(lrres2, ctrline=list(gamma=0.95, col="red", lty="dashed"))
# goodness-of-fit plot for items 1, 14, 24, and 25
# additional 95 percent confidence ellipses, default style
plotGOF(lrres2, beta.subset=c(14,25,24,1), conf=list())
+
+# goodness-of-fit plot for items 1, 14, 24, and 25
+# for items 1 and 24 additional 95 percent confidence ellipses
+# using colours for these 2 items from the colorspace package
+\dontrun{
+library(colorspace)
+colors<-rainbow_hcl(2)
+plotGOF(lrres2, beta.subset=c(14,25,24,1), conf=list(which=c(1,14), col=colors))
+}
}
\keyword{models}
diff --git a/man/MLoef.Rd b/man/MLoef.Rd
old mode 100644
new mode 100755
index 74196fb..33b3a2b
--- a/man/MLoef.Rd
+++ b/man/MLoef.Rd
@@ -9,11 +9,11 @@ MLoef(robj, splitcr = "median")
}
\arguments{
\item{robj}{Object of class \code{Rm}.}
- \item{splitcr}{Split criterion to define two groups of item.
+ \item{splitcr}{Split criterion to define the item groups.
\code{"median"} and \code{"mean"} split items in two groups based on their
items' raw scores. \code{splitcr} can also be a vector of length k (where k
- denotes the number of items) that takes two distinct values to define groups
- used for the Martin-Loef Test.}
+ denotes the number of items) that takes two or more distinct values to
+ define groups used for the Martin-Loef Test.}
}
\details{
This function implements a generalization of the Martin-Loef test for polytomous
@@ -23,12 +23,15 @@ MLoef(robj, splitcr = "median")
% least 2 valid responses in each group of items.
If the split criterion is \code{"median"} or \code{"mean"} and one or more items'
- raw scores are equal the median resp. mean, \code{MLoef} will issue a warning
- that those items are assigned to the lower raw score group. \code{summary.MLoef}
- gives detailed information about the allocation of all items.
+ raw scores are equal the median resp. mean, \code{MLoef} will assign those items
+ to the lower raw score group. \code{summary.MLoef} gives detailed information
+ about the allocation of all items.
\code{summary} and \code{print} methods are available for objects of class
\code{MLoef}.
+
+ An 'exaxt' version of the Martin-Loef test for binary items is implemented
+ in the function \code{\link{NPtest}}.
}
\value{
\code{MLoef} returns an object of class \code{MLoef} containing:
@@ -56,9 +59,9 @@ Rost, J. (2004). \emph{Lehrbuch Testtheorie -- Testkonstruktion.} Bern: Huber.
%\note{}
\seealso{\code{\link{LRtest}}, \code{\link{Waldtest}}}
\examples{
-# Martin-Loef-test on dichotomous Rasch model using "median"
-# and a user-defined split
-splitvec <- c(1, 1, 1, 1, 0, 1, 0, 0, 1, 0)
+# Martin-Loef-test on dichotomous Rasch model using "median" and a user-defined
+# split vector. Note that group indicators can be of character and/or numeric.
+splitvec <- c(1, 1, 1, "x", "x", "x", 0, 0, 1, 0)
res <- RM(raschdat1[,1:10])
diff --git a/man/NPtest.Rd b/man/NPtest.Rd
old mode 100644
new mode 100755
index 9625297..248187a
--- a/man/NPtest.Rd
+++ b/man/NPtest.Rd
@@ -2,7 +2,7 @@
\Rdversion{1.1}
\alias{NPtest}
\title{function to perform nonparametric Rasch model tests}
-\description{A variety of nonparametric tests as proposed by Ponocny(2001) are implemented. The function operates on
+\description{A variety of nonparametric tests as proposed by Ponocny(2001) and an 'exact' version of the Martin-Loef test are implemented. The function operates on
random binary matrices that have been generated using an
MCMC algorithm (Verhelst, 2008) from the RaschSampler package (Hatzinger, Mair, and Verhelst, 2009).
}
@@ -73,12 +73,17 @@ NPtest(obj, n=NULL, method = "T1", ...)
Gobal test for local dependence. The statistic calculates the sum of absolute deviations between the observed inter-item correlations
and the expected correlations.
}
+
+ The 'exact' version of the \bold{Martin-Loef} statistic is specified via \code{method = "MLoef"} and optionally \code{splitcr}
+ (see \code{\link{MLoef}}).
}
\value{
Depends on the method used. For each method a list is returned. The returned objects are of class
\code{T1obj}, \code{T2obj}, \code{T4obj}, \code{T7obj}, \code{T7aobj}, \code{T10obj}, \code{T11obj} corresponding to the method used.
The main output element is \code{prop} giving the one-sided p-value, i.e., the number of statistics from the sampled matrices which are equal
or exceed the statistic based on the observed data. For \emph{T1} and \emph{T7a} \code{prop} is a vector.
+For the \emph{Martin-Loef} test the returned object is of class \code{MLobj}. Besides other elements, it contains a \code{prop} vector and \code{MLres}, the output
+object from the asymptotic Martin-Loef test on the input data.
}
\references{
Ponocny, I. (2001) Nonparametric goodness-of-fit tests for the rasch model. Psychometrika, Volume 66, Number 3\cr
@@ -161,7 +166,15 @@ t103
##---- T11 ------------------------------------------------------
t11<-NPtest(rmat,method="T11")
t11
+
+##---- Martin-Loef ----------------------------------------------
+\dontrun{
+# takes a while ...
+data(raschdat1)
+split<-rep(1:3, each=10)
+NPtest(raschdat1, n=100, method="MLoef", splitcr=split)
}
+}
\keyword{htest}
\keyword{nonparametric}
diff --git a/man/PCM.Rd b/man/PCM.Rd
old mode 100644
new mode 100755
diff --git a/man/RM.Rd b/man/RM.Rd
old mode 100644
new mode 100755
diff --git a/man/RSM.Rd b/man/RSM.Rd
old mode 100644
new mode 100755
diff --git a/man/Waldtest.Rd b/man/Waldtest.Rd
old mode 100644
new mode 100755
diff --git a/man/anova.llra.Rd b/man/anova.llra.Rd
new file mode 100755
index 0000000..cf039c4
--- /dev/null
+++ b/man/anova.llra.Rd
@@ -0,0 +1,65 @@
+\name{anova.llra}
+\alias{anova.llra}
+\alias{anova.llra.default}
+
+
+\title{Analysis of Deviance for Linear Logistic Models with Relaxed Assumptions
+}
+\description{Compute an analysis of deviance table for one or more LLRA.
+}
+\usage{
+\method{anova}{llra}(object, ...)
+}
+\arguments{
+ \item{object, ... }{objects of class "llra", typically the result of a
+ call to \code{\link{LLRA}}.
+ }
+}
+\details{
+An analysis of deviance table will be calculated. The models in rows are
+ordered from the smallest to the largest model. Each row shows the
+number of parameters (Npar) and the log-likelihood (logLik). For all but
+the first model, the parameter difference (df) and the difference in
+deviance or the likelihood ratio (-2LR) is given between two subsequent
+models (with increasing complexity). Please note that interpreting these
+values only makes sense if the models are nested.
+
+The table also contains p-values comparing the reduction in the
+deviance to the df for each row based on the asymptotic Chi^2-Distribution of the Likelihood ratio test statistic.
+}
+\value{
+An object of class \code{"anova"} inheriting from class \code{"data.frame"}.
+}
+\author{
+Thomas Rusch
+}
+\section{Warning:}{
+The comparison between two or more models by \code{anova} will only be valid
+if they are fitted to the same dataset and if the models are nested. The
+function does not check if that is the case.
+}
+\seealso{
+The model fitting function \code{\link{LLRA}}.
+}
+\examples{
+ \dontrun{
+ ##An LLRA with 2 treatment groups and 1 baseline group, 5 items and 4
+ ##time points. Item 1 is dichotomous, all others have 3, 4, 5, 6
+ ##categories respectively.
+ data("llraDat2")
+
+ #fit LLRA
+ ex2 <- LLRA(llraDat2[,1:20],mpoints=4,groups=llraDat2[,21])
+
+ #Imposing a linear trend for items 2 and 3 using collapse_W
+ collItems2 <- list(c(32,37,42),c(33,38,43))
+ newNames2 <- c("trend.I2","trend.I3")
+ Wnew <- collapse_W(ex2$W,collItems2,newNames2)
+
+ #Estimating LLRA with the linear trend for item 2 and 3
+ ex2new <- LLRA(llraDat2[1:20],W=Wnew,mpoints=4,groups=llraDat2[21])
+
+ #comparing models with likelihood ratio test
+ anova(ex2,ex2new)
+}
+}
diff --git a/man/build_W.Rd b/man/build_W.Rd
new file mode 100755
index 0000000..120f96e
--- /dev/null
+++ b/man/build_W.Rd
@@ -0,0 +1,90 @@
+\name{build_W}
+\alias{build_W}
+\alias{build_catdes}
+\alias{build_trdes}
+\alias{build_effdes}
+\alias{get_item_cats}
+
+\title{
+Automatized Construction of LLRA Design Matrix
+}
+\description{
+Builds a design matrix for LLRA from scratch.
+}
+\usage{
+build_W(X, nitems, mpoints, grp_n, groupvec, itmgrps)
+}
+\arguments{
+ \item{X}{Data matrix as described in Hatzinger and Rusch (2009). It
+ must be of long format, e.g. for each person all item answers are written in subsequent rows. The columns correspond to time
+ points. Missing values are not allowed. It can easily be
+ constructed from data in wide format with
+ \code{matrix(unlist(data),ncol=mpoints)} or from \code{\link{llra.datprep}}.
+ }
+ \item{nitems}{The number of items.
+}
+ \item{mpoints}{The number of time points.
+}
+ \item{grp_n}{A vector of number of subjects per g+1 groups (e.g. g
+ treatment or covariate groups and 1 control or baseline group.
+ The sizes must be ordered like the corresponding groups.
+ }
+ \item{groupvec}{Assignment vector, i.e. which person belongs to which
+ treatment/item group
+ }
+ \item{itmgrps}{Specifies how many groups of items there are.
+ }
+}
+\details{
+The function is designed to be modular and calls four internal function
+\code{build_effdes} (for treatment/covariate effects), \code{build_trdes} (for trend
+effects), \code{build_catdes} (for category parameter design matrix) and
+\code{get_item_cats} (checks how many categories each item has). Those functions are not intended to be used by the user.
+
+Labeling of effects also happens in the internal functions.
+}
+\value{An LLRA design matrix as described by Hatzinger and Rusch
+ (2009). This can be passed as the \code{W} argument to \code{LLRA} or
+ \code{LPCM}.
+
+ The design matrix specifies every item to lie on its own
+ dimension. Hence at every time point > 1, there are effects for
+ each treatment or covariate group as well as trend effects for every
+ item. Therefore overall there are items x (groups-1) x (time points-1)
+ covariate effect parameters and items x (time points-1) trend
+ parameters specified. For polytomous items there also are parameters
+ for each category with the first and second category being equated for each item. They
+ need not be equidistant. The number of parameters therefore increase
+ quite rapidly for any additional time point, item or covariate group.
+}
+\references{
+Hatzinger, R. and Rusch, T. (2009) IRT models with relaxed assumptions
+in eRm: A manual-like instruction. \emph{Psychology Science Quarterly},
+ \bold{51}, pp. 87--120, \url{http://erm.r-forge.r-project.org/psq_1_2009_06_87-120.pdf}
+}
+\author{
+Thomas Rusch
+}
+\section{Warning }{A warning is printed that the first two categories
+ for polytomous items are equated.}
+
+\seealso{
+This function is used for automatic generation of the design matrix in \code{\link{LLRA}}.
+}
+\examples{
+ ##An LLRA with 2 treatment groups and 1 baseline group, 5 items and 4
+ ##time points. Item 1 is dichotomous, all others have 3, 4, 5, 6
+ ##categories respectively.
+ data("llraDat2")
+ llraDat2a <- matrix(unlist(llraDat2[1:20]),ncol=4)
+ groupvec <-rep(1:3*5,each=20)
+ W <- build_W(llraDat2a,nitems=5,mpoints=4,grp_n=c(10,20,40),groupvec=groupvec,itmgrps=1:5)
+
+ #There are 55 parameters
+ dim(W)
+
+ #Estimating LLRA by specifiying W
+ \dontrun{
+ ex2W <- LLRA(llraDat2[1:20],W=W,mpoints=4,groups=llraDat2[21])
+ }
+}
\ No newline at end of file
diff --git a/man/collapse_W.Rd b/man/collapse_W.Rd
new file mode 100755
index 0000000..9106779
--- /dev/null
+++ b/man/collapse_W.Rd
@@ -0,0 +1,76 @@
+\name{collapse_W}
+\alias{collapse_W}
+
+\title{
+Convenient Collapsing of LLRA Design Matrix
+}
+\description{
+Collapses columns of a design matrix for LLRA to specify different
+parameter restrictions in \code{LLRA}.
+}
+\usage{
+collapse_W(W, listItems, newNames)
+}
+\arguments{
+ \item{W}{A design matrix (for LLRA), typically from a call to
+ \code{\link{build_W}} or component \code{$W} from \code{\link{LLRA}}
+ or \code{\link{LPCM}}
+ }
+ \item{listItems}{A list of numeric vectors. Each component of the list specifies
+ columns to be collapsed together.
+ }
+ \item{newNames}{An (optional) character vector specifying the names of
+ the collapsed effects.
+ }
+}
+\details{
+This function is a convenience function to collapse a design matrix,
+i.e. to specify linear trend or treatment effects and so on. Collapsing
+here means that effects in columns are summed up. For this, a list of numeric
+vectors with the column indices of columns to be collapsed have to be
+passed to the function. For example, if you want to collapse column 3, 6
+and 8 into one new effect and 1, 4 and 9 into another it needs to be
+passed with \code{list(c(3,6,8),c(1,4,9))}.
+
+The new effects can be given names by passing a character vector to the
+function with equal length as the list.
+}
+
+\value{An LLRA design matrix as described by Hatzinger and Rusch
+ (2009). This can be passed as the \code{W} argument to \code{LLRA} or
+ \code{LPCM}.
+}
+\references{
+Hatzinger, R. and Rusch, T. (2009) IRT models with relaxed assumptions
+in eRm: A manual-like instruction. \emph{Psychology Science Quarterly},
+ \bold{51}, pp. 87--120, \url{http://erm.r-forge.r-project.org/psq_1_2009_06_87-120.pdf}
+}
+\author{
+Thomas Rusch
+}
+\seealso{
+The function to build design matrices from scratch, \code{\link{build_W}}.
+}
+\examples{
+ ##An LLRA with 2 treatment groups and 1 baseline group, 5 items and 4
+ ##time points. Item 1 is dichotomous, all others have 3, 4, 5, 6
+ ##categories respectively.
+ data("llraDat2")
+ llraDat2a <- matrix(unlist(llraDat2[1:20]),ncol=4)
+ groupvec <-rep(1:3*5,each=20)
+ W <- build_W(llraDat2a, nitems=5, mpoints=4, grp_n=c(10,20,40), groupvec=groupvec,
+ itmgrps=1:5)
+
+ #There are 55 parameters to be estimated
+ dim(W)
+
+ #Imposing a linear trend for the second item ,i.e. parameters in
+ #columns 32, 37 and 42 need to be
+ #collapsed into a single column.
+ collItems1 <- list(c(32,37,42))
+ newNames1 <- c("trend.I2")
+ Wstar1 <- collapse_W(W,collItems1)
+
+ #53 parameters need to be estimated
+ dim(Wstar1)
+}
\ No newline at end of file
diff --git a/man/eRm-package.Rd b/man/eRm-package.Rd
old mode 100644
new mode 100755
index 77b18fc..85d6fbf
--- a/man/eRm-package.Rd
+++ b/man/eRm-package.Rd
@@ -20,8 +20,8 @@ http://r-forge.r-project.org/projects/erm/.
\tabular{ll}{
Package: \tab eRm\cr
Type: \tab Package\cr
-Version: \tab 0.13-4\cr
-Date: \tab 2011-03-23\cr
+Version: \tab 0.14-0\cr
+Date: \tab 2011-06-05\cr
License: \tab GPL\cr
}
The basic input units for the functions are the person-item matrix X and the design matrix W.
@@ -34,19 +34,24 @@ The linear extensions provide the possibility to fit a more restricted model tha
such as \code{LLTM(X, W)}, \code{LRSM(X, W)},\code{LPCM(X, W)}, but
also a generalization by imposing repeated measurement designs and group contrasts. These models can
be estimated by using, e.g.,
-\code{LLTM(X, W, mpoints = 2, groupvec = G)},\cr
-\code{LRSM(X, W, mpoints = 2, groupvec = G)}, and\cr
-\code{LPCM(X, W, mpoints = 2, groupvec = G)}. \cr
+\code{LLTM(X, W, mpoints = 2, groupvec = g)},\cr
+\code{LRSM(X, W, mpoints = 2, groupvec = g)},\cr
+\code{LPCM(X, W, mpoints = 2, groupvec = g)},\cr
+and as very flexible multidimensional model for repeated measurements
+\code{LLRA(X, W, mpoints = 2, groups = G)},\cr
\code{mpoints} specifies the number of measurement or time points,
-\code{G} is a vector with the group membership for each subject
-ordered according to the rows of the data matrix.
+\code{g} is a vector with the group membership for each subject,
+ordered according to the rows of the data matrix, and
+\code{G} is a matrix with subject covariates (e.g., treatments),
-\code{RM} produces an object belonging to the classes \code{dRM}, \code{RM}, and
+\code{RM} produces an object belonging to the classes \code{dRm}, \code{Rm}, and
\code{eRm}. \code{PCM} and \code{RSM} produce objects belonging to the classes
-\code{RM} and \code{eRm}, whereas results of \code{LLTM}, \code{LRSM}, and \code{LLTM} are object of class \code{eRm}.
+\code{Rm} and \code{eRm}, whereas results of \code{LLTM}, \code{LRSM}, \code{LLTM}
+and \code{LLRA} are objects of class \code{eRm}. For a detailled overview of all
+classes defined in the package and the functions depending on them see the package's vignette.
-We acknowledge Julian Gilbey for writing the \code{plotPWmap} function and Kathrin Gruber
-for the function \code{plotDIF}.
+We acknowledge Julian Gilbey for writing the \code{plotPWmap} function, Kathrin Gruber
+for the function \code{plotDIF}, and Thomas Rusch for \code{LLRA} and related utilities.
The \code{eRm} package contains functions from the packages \code{sna}, \code{gtools} and \code{ROCR}.
Thanks to Carter T. Butts, Gregory R. Warnes, and Tobias Sing et al.
}
@@ -55,7 +60,7 @@ Thanks to Carter T. Butts, Gregory R. Warnes, and Tobias Sing et al.
\code{fitctrl <- "nlm"} or \code{fitctrl <- "optim"}.}
-\author{Patrick Mair, Reinhold Hatzinger, Marco Maier, Julian Gilbey
+\author{Patrick Mair, Reinhold Hatzinger, Marco Maier, and others
Maintainer: Patrick Mair <patrick.mair at wu.ac.at>
}
diff --git a/man/gofIRT.Rd b/man/gofIRT.Rd
old mode 100644
new mode 100755
diff --git a/man/itemfit.ppar.Rd b/man/itemfit.ppar.Rd
old mode 100644
new mode 100755
diff --git a/man/llra.datprep.Rd b/man/llra.datprep.Rd
new file mode 100755
index 0000000..27714b0
--- /dev/null
+++ b/man/llra.datprep.Rd
@@ -0,0 +1,70 @@
+\name{llra.datprep}
+\alias{llra.datprep}
+
+\title{Prepare Data Set for LLRA Analysis
+}
+\description{
+Converts wide data matrix in long format, sorts subjects according to
+groups and builds assigment vector.
+}
+\usage{
+llra.datprep(X, mpoints, groups, baseline)
+}
+\arguments{
+ \item{X}{Data matrix as described in Hatzinger and Rusch (2009). It
+ must be of wide format, e.g. for each person all item answers are
+ written in columns for t1, t2, t3 etc. Hence each row corresponds to
+ all observations for a single person.
+ Missing values are not allowed.
+ }
+ \item{mpoints}{The number of time points.
+ }
+ \item{groups}{Vector, matrix or data frame with subject/treatment
+ covariates.
+ }
+ \item{baseline}{An optional vector with the baseline values for the
+ columns in group.}
+}
+\details{The function converts a data matrix from wide to long fromat as
+ needed for LLRA. Additionally it sorts the subjects according to the
+ different treatment/covariate groups. The group with the lowest
+ (alpha-)numerical value will be the
+ baseline.
+
+ Treatment and covariate groups are either defined by a vector, or by a
+ matrix or data frame. The latter will be combined to a vector of
+ groups corresponding to a combination of each factor level per column
+ with the factor levels of the other column. The (constructed or
+ passed) vector will then be used to create the assignment vector.
+}
+\value{
+Returns a list with the components
+\item{X}{Data matrix in long format with subjects sorted by groups.}
+\item{assign.vec}{The assignment vector.}
+\item{grp_n}{A vector of the number of subjects in each group.}
+}
+\author{
+Reinhold Hatzinger
+}
+\seealso{
+The function that uses this is \code{\link{LLRA}}. The values from
+ \code{llra.datprep} can be passed to \code{\link{build_W}}.
+}
+\examples{
+ # example 3 items, 3 timepoints, n=10, 2x2 treatments
+ dat<-sim.rasch(10,9)
+ tr1<-sample(c("a","b"),10,r=TRUE)
+ tr2<-sample(c("x","y"),10,r=TRUE)
+
+ # one treatment
+ res<-llra.datprep(dat,mpoints=3,groups=tr1)
+ res<-llra.datprep(dat,mpoints=3,groups=tr1,baseline="b")
+
+ # two treatments
+ res<-llra.datprep(dat,mpoints=3,groups=cbind(tr1,tr2))
+ res<-llra.datprep(dat,mpoints=3,groups=cbind(tr1,tr2),baseline=c("b","x"))
+
+ # two treatments - data frame
+ tr.dfr<-data.frame(tr1, tr2)
+ res<-llra.datprep(dat,mpoints=3,groups=tr.dfr)
+}
diff --git a/man/llraDat1.Rd b/man/llraDat1.Rd
new file mode 100755
index 0000000..1ae4167
--- /dev/null
+++ b/man/llraDat1.Rd
@@ -0,0 +1,61 @@
+\name{llraDat1}
+\alias{llraDat1}
+\docType{data}
+\title{An Artifical LLRA Data Set
+}
+\description{
+Artificial data set of 5 items, 5 time points and 5 groups for LLRA.
+}
+\usage{data(llraDat1)}
+\format{
+ A data frame with 150 observations of 26 variables.
+ \itemize{
+ \item{t1.I1}{ Answers to item 1 at time point 1}
+ \item{t1.I2}{ Answers to item 2 at time point 1}
+ \item{t1.I3}{ Answers to item 3 at time point 1}
+ \item{t1.I4}{ Answers to item 4 at time point 1}
+ \item{t1.I5}{ Answers to item 5 at time point 1}
+ \item{t2.I1}{ Answers to item 1 at time point 2}
+ \item{t2.I2}{ Answers to item 2 at time point 2}
+ \item{t2.I3}{ Answers to item 3 at time point 2}
+ \item{t2.I4}{ Answers to item 4 at time point 2}
+ \item{t2.I5}{ Answers to item 5 at time point 2}
+ \item{t3.I1}{ Answers to item 1 at time point 3}
+ \item{t3.I2}{ Answers to item 2 at time point 3}
+ \item{t3.I3}{ Answers to item 3 at time point 3}
+ \item{t3.I4}{ Answers to item 4 at time point 3}
+ \item{t3.I5}{ Answers to item 5 at time point 3}
+ \item{t4.I1}{ Answers to item 1 at time point 4}
+ \item{t4.I2}{ Answers to item 2 at time point 4}
+ \item{t4.I3}{ Answers to item 3 at time point 4}
+ \item{t4.I4}{ Answers to item 4 at time point 4}
+ \item{t4.I5}{ Answers to item 5 at time point 4}
+ \item{t5.I1}{ Answers to item 1 at time point 5}
+ \item{t5.I2}{ Answers to item 2 at time point 5}
+ \item{t5.I3}{ Answers to item 3 at time point 5}
+ \item{t5.I4}{ Answers to item 4 at time point 5}
+ \item{t5.I5}{ Answers to item 5 at time point 5}
+ \item{groups}{ The group membership}
+}
+}
+\details{
+This is a data set as described in Hatzinger and Rusch (2009). 5 items
+were measured at 5 time points (in columns). Each row corresponds to one
+person (P1 to P150).
+There are 4 treatment groups and a control group. Treatment group G5 has
+size 10 (the first ten subjects),
+treatment group G4 has size 20, treatment group G3 has size 30, treatment
+group G2 has size 40 and the control group CG has size 50 (the last 50
+subjects). Item 1 is dichotomous, all others are polytomous. Item 2, 3, 4 and 5 have 3, 4, 5, 6 categories
+respectively.
+}
+\references{
+Hatzinger, R. and Rusch, T. (2009) IRT models with relaxed assumptions
+in eRm: A manual-like instruction. \emph{Psychology Science Quarterly}, \bold{51},
+pp. 87--120,
+\url{http://erm.r-forge.r-project.org/psq_1_2009_06_87-120.pdf}
+}
+\examples{
+data(llraDat1)
+}
+\keyword{datasets}
diff --git a/man/llraDat2.Rd b/man/llraDat2.Rd
new file mode 100755
index 0000000..1e0a970
--- /dev/null
+++ b/man/llraDat2.Rd
@@ -0,0 +1,54 @@
+\name{llraDat2}
+\alias{llraDat2}
+\docType{data}
+\title{An Artifical LLRA Data Set
+}
+\description{
+Artificial data set of 70 subjects with 5 items, 4 time points and 3 groups for LLRA.
+}
+\usage{data(llraDat2)}
+\format{
+ A data frame with 70 observations of 21 variables.
+ \itemize{
+ \item{t1.I1}{ Answers to item 1 at time point 1}
+ \item{t1.I2}{ Answers to item 2 at time point 1}
+ \item{t1.I3}{ Answers to item 3 at time point 1}
+ \item{t1.I4}{ Answers to item 4 at time point 1}
+ \item{t1.I5}{ Answers to item 5 at time point 1}
+ \item{t2.I1}{ Answers to item 1 at time point 2}
+ \item{t2.I2}{ Answers to item 2 at time point 2}
+ \item{t2.I3}{ Answers to item 3 at time point 2}
+ \item{t2.I4}{ Answers to item 4 at time point 2}
+ \item{t2.I5}{ Answers to item 5 at time point 2}
+ \item{t3.I1}{ Answers to item 1 at time point 3}
+ \item{t3.I2}{ Answers to item 2 at time point 3}
+ \item{t3.I3}{ Answers to item 3 at time point 3}
+ \item{t3.I4}{ Answers to item 4 at time point 3}
+ \item{t3.I5}{ Answers to item 5 at time point 3}
+ \item{t4.I1}{ Answers to item 1 at time point 4}
+ \item{t4.I2}{ Answers to item 2 at time point 4}
+ \item{t4.I3}{ Answers to item 3 at time point 4}
+ \item{t4.I4}{ Answers to item 4 at time point 4}
+ \item{t4.I5}{ Answers to item 5 at time point 4}
+ \item{groups}{ The group membership}
+}
+}
+\details{
+This is a data set as described in Hatzinger and Rusch (2009). 5 items
+were measured at 4 time points (in columns). Each persons answers to the
+items are recorded in the rows. There are 2
+treatment groups and a control group. Treatment group 2 has size, 10,
+treatment group 1 has size 20 and the control group has size 40. Item 1 is dichotomous, all others
+are polytomous. Item 2, 3, 4 and 5 have 3, 4, 5, 6 categories
+respectively.
+}
+\references{
+Hatzinger, R. and Rusch, T. (2009) IRT models with relaxed assumptions
+in eRm: A manual-like instruction. \emph{Psychology Science Quarterly}, \bold{51},
+pp. 87--120,
+\url{http://erm.r-forge.r-project.org/psq_1_2009_06_87-120.pdf}
+}
+\examples{
+data(llraDat2)
+}
+\keyword{datasets}
diff --git a/man/llradat3.Rd b/man/llradat3.Rd
new file mode 100755
index 0000000..26cbf1d
--- /dev/null
+++ b/man/llradat3.Rd
@@ -0,0 +1,34 @@
+\name{llradat3}
+\alias{llradat3}
+\docType{data}
+\title{An Artifical LLRA Data Set
+}
+\description{
+Artificial data set of 3 items, 2 time points and 2 groups for LLRA. It
+is example 6 from Hatzinger and Rusch (2009).
+}
+\usage{data(llradat3)}
+\format{
+ A data frame with 60 observations of 6 variables.
+ \itemize{
+ \item{V1}{ Answers to item 1 at time point 1}
+ \item{V2}{ Answers to item 2 at time point 1}
+ \item{V3}{ Answers to item 3 at time point 1}
+ \item{V4}{ Answers to item 1 at time point 2}
+ \item{V5}{ Answers to item 2 at time point 2}
+ \item{V6}{ Answers to item 3 at time point 2}
+}
+}
+\details{
+This is a data set as described in Hatzinger and Rusch (2009).
+}
+\references{
+Hatzinger, R. and Rusch, T. (2009) IRT models with relaxed assumptions
+in eRm: A manual-like instruction. \emph{Psychology Science Quarterly}, \bold{51},
+pp. 87--120,
+\url{http://erm.r-forge.r-project.org/psq_1_2009_06_87-120.pdf}
+}
+\examples{
+data(llradat3)
+}
+\keyword{datasets}
diff --git a/man/person.parameter.Rd b/man/person.parameter.Rd
old mode 100644
new mode 100755
diff --git a/man/plotDIF.Rd b/man/plotDIF.Rd
old mode 100644
new mode 100755
index 140abf6..5f09aef
--- a/man/plotDIF.Rd
+++ b/man/plotDIF.Rd
@@ -9,7 +9,7 @@ Performs an plot of item parameter conficence intervals based on \code{LRtest} s
\usage{
plotDIF(object, item.subset = NULL, gamma = 0.95, main = NULL,
xlim = NULL, xlab = " ", ylab=" ", col = NULL,
- distance = 10, splitnames=NULL, leg = FALSE, legpos="bottomleft", ...)
+ distance, splitnames=NULL, leg = FALSE, legpos="bottomleft", ...)
}
\arguments{
@@ -41,7 +41,8 @@ By default the color for the drawn confidence lines is determined automatically
is depicted in the same color.
}
\item{distance}{
-Distance between the drawn confidence lines, default is division by factor 10 .
+Distance between each item's confidence lines -- if omitted, the distance shrinks with increasing numbers of split criteria. Can be overriden using values in (0, 0.5).
+%Distance between the drawn confidence lines, default is division by factor 10 .
}
\item{splitnames}{
For labeling the splitobjects in the legend (returns a nicer output).
diff --git a/man/plotGR.Rd b/man/plotGR.Rd
new file mode 100755
index 0000000..98307c3
--- /dev/null
+++ b/man/plotGR.Rd
@@ -0,0 +1,57 @@
+\name{plotGR}
+\alias{plotGR}
+
+\title{Plot Treatment or Covariate Effects for LLRA
+}
+\description{
+Plots treatment or covariate group effects over time.
+}
+\usage{
+plotGR(object, ...)
+}
+\arguments{
+ \item{object}{an object of class "llra".
+ }
+ \item{\dots}{Additional parameters to be passed to and from other
+ methods.
+ }
+}
+\details{
+The plot is a lattice plot with each panel corresponding to an item. The
+effects are plotted for each groups (including baseline) over the
+different time points. The groups are given the same names as for the
+parameter estimates (derived from groupvec).
+
+Please note that all effects are to be interpreted relative to the
+baseline.
+
+Currently, this function only works for a full item x treatment x
+timepoints LLRA. Collapsed effects will not be displayed properly.
+}
+\author{
+Thomas Rusch
+}
+\seealso{
+The plot method for trend effects \code{\link{plotTR}}.
+}
+\section{Warning:}{
+ Objects of class \code{"llra"} that contain estimates from a collapsed
+ data matrix will not be displayed correctly.
+}
+\examples{
+ ##Example 6 from Hatzinger & Rusch (2009)
+ data("llradat3")
+ groups <- c(rep("TG",30),rep("CG",30))
+ llra1 <- LLRA(llradat3,mpoints=2,groups=groups)
+ summary(llra1)
+ plotGR(llra1)
+
+ ##An LLRA with 2 treatment groups and 1 baseline group, 5 items and 4
+ ##time points. Item 1 is dichotomous, all others have 3, 4, 5, 6
+ ##categories respectively.
+ \dontrun{
+ data("llraDat2")
+ ex2 <- LLRA(llraDat2[1:20],mpoints=4,groups=llraDat2[21])
+ plotGR(ex2)
+ }
+}
\ No newline at end of file
diff --git a/man/plotICC.Rd b/man/plotICC.Rd
old mode 100644
new mode 100755
diff --git a/man/plotPImap.Rd b/man/plotPImap.Rd
old mode 100644
new mode 100755
diff --git a/man/plotPWmap.Rd b/man/plotPWmap.Rd
old mode 100644
new mode 100755
diff --git a/man/plotTR.Rd b/man/plotTR.Rd
new file mode 100755
index 0000000..011a2c1
--- /dev/null
+++ b/man/plotTR.Rd
@@ -0,0 +1,54 @@
+\name{plotTR}
+\alias{plotTR}
+
+\title{Plot Trend Effects for LLRA
+}
+\description{
+Plots trend effects over time.
+}
+\usage{
+plotTR(object, ...)
+}
+\arguments{
+ \item{object}{an object of class \code{"llra"}
+}
+ \item{\dots}{Additional parameters to be passed to and from other methods}
+}
+\details{
+ The plot is a lattice plot with one panel. The effects for each items
+ are plotted over the different time points.
+
+ Please
+ note that all effects are to be interpreted relative to the baseline
+ (i.e. t1).
+
+ Currently, this function only works for a full item x treatment x
+ timepoints LLRA. Collapsed effects will not be displayed properly.
+}
+\author{
+Thomas Rusch
+}
+\seealso{
+ The plot method for treatment effects \code{"plotGR"}.
+}
+\section{Warning:}{
+ Objects of class \code{"llra"} that contain estimates from a collapsed
+ data matrix will not be displayed correctly.
+}
+\examples{
+ ##Example 6 from Hatzinger & Rusch (2009)
+ data("llradat3")
+ groups <- c(rep("TG",30),rep("CG",30))
+ llra1 <- LLRA(llradat3,mpoints=2,groups=groups)
+ summary(llra1)
+ plotTR(llra1)
+
+ ##An LLRA with 2 treatment groups and 1 baseline group, 5 items and 4
+ ##time points. Item 1 is dichotomous, all others have 3, 4, 5, 6
+ ##categories respectively.
+ \dontrun{
+ data("llraDat2")
+ ex2 <- LLRA(llraDat2[1:20],mpoints=4,groups=llraDat2[21])
+ plotTR(ex2)
+ }
+}
diff --git a/man/predict.ppar.Rd b/man/predict.ppar.Rd
old mode 100644
new mode 100755
diff --git a/man/print.eRm.Rd b/man/print.eRm.Rd
old mode 100644
new mode 100755
diff --git a/man/raschdat.Rd b/man/raschdat.Rd
old mode 100644
new mode 100755
diff --git a/man/sim.2pl.Rd b/man/sim.2pl.Rd
old mode 100644
new mode 100755
diff --git a/man/sim.locdep.Rd b/man/sim.locdep.Rd
old mode 100644
new mode 100755
diff --git a/man/sim.rasch.Rd b/man/sim.rasch.Rd
old mode 100644
new mode 100755
diff --git a/man/sim.xdim.Rd b/man/sim.xdim.Rd
old mode 100644
new mode 100755
diff --git a/man/stepwiseIt.Rd b/man/stepwiseIt.Rd
old mode 100644
new mode 100755
diff --git a/man/summary.llra.Rd b/man/summary.llra.Rd
new file mode 100755
index 0000000..cd7eb00
--- /dev/null
+++ b/man/summary.llra.Rd
@@ -0,0 +1,69 @@
+\name{summary.llra}
+\alias{summary.llra}
+\alias{print.summary.llra}
+
+\title{Summarizing Linear Logistic Models with Relaxed Assumptions (LLRA)
+}
+\description{
+\code{summary} method for class \code{"llra"}
+}
+\usage{
+\method{summary}{llra}(object, gamma, ...)
+
+\method{print}{summary.llra}(x, ...)
+}
+\arguments{
+ \item{object}{an object of class "llra", typically result of a call to
+ \code{\link{LLRA}}.
+ }
+ \item{x}{an object of class "summary.llra", usually, a result of a call
+ to \code{summary.llra}.
+ }
+ \item{gamma}{The level of confidence for the confidence
+ intervals. Default is 0.95.}
+ \item{\dots}{further arguments passed to or from other methods.
+ }
+}
+\details{
+Objects of class \code{"summary.llra"} contain all parameters of interest plus the confidence intervals.
+
+\code{print.summary.llra} rounds the values to 3 digits and displays
+them nicely.
+}
+\value{
+ The function \code{summary.lllra} computes and returns a list of
+ summary statistics of the fitted LLRA given in object, reusing the
+ components (list elements) \code{call}, \code{etapar},
+ \code{iter}, \code{loglik}, \code{model}, \code{npar} and \code{se.etapar} from its argument, plus
+
+ \item{ci}{The upper and lower confidence interval borders.}
+}
+\author{
+Thomas Rusch
+}
+
+\seealso{
+The model fitting function \code{\link{LLRA}}.
+}
+\examples{
+ ##Example 6 from Hatzinger & Rusch (2009)
+ data("llradat3")
+ groups <- c(rep("TG",30),rep("CG",30))
+ llra1 <- LLRA(llradat3,mpoints=2,groups=groups)
+ summary(llra1)
+
+ ##An LLRA with 2 treatment groups and 1 baseline group, 5 items and 4
+ ##time points. Item 1 is dichotomous, all others have 3, 4, 5, 6
+ ##categories respectively.
+ \dontrun{
+ data("llraDat2")
+ ex2 <- LLRA(llraDat2[1:20],mpoints=4,llraDat2[21])
+ sumEx2 <- summary(ex2, gamma=0.95)
+
+ #print the summary
+ sumEx2
+
+ #get confidence intervals
+ sumEx2$ci
+}
+}
diff --git a/man/thresholds.Rd b/man/thresholds.Rd
old mode 100644
new mode 100755
diff --git a/src/components.c b/src/components.c
old mode 100644
new mode 100755
diff --git a/src/components.h b/src/components.h
old mode 100644
new mode 100755
diff --git a/src/geodist.c b/src/geodist.c
old mode 100644
new mode 100755
diff --git a/src/geodist.h b/src/geodist.h
old mode 100644
new mode 100755
--
Alioth's /usr/local/bin/git-commit-notice on /srv/git.debian.org/git/debian-science/packages/r-cran-erm.git
More information about the debian-science-commits
mailing list