Dear Alexander,
I have problems with standard error estimation via tam.se() for models that were estimated using QMC, see example below. I don't know if this is a feature or a bug. But in case it's a bug, I propose the committed changes. I tried a few examples and compared the SEs for models with and without QMC, and they were comparable IMHO. Feel free to adapt this to your needs.
Best, Hansjörg
library("TAM")
#> Loading required package: CDM
#> Loading required package: mvtnorm
#> **********************************
#> ** CDM 7.3-17 (2019-03-18 18:33:40)
#> ** Cognitive Diagnostic Models **
#> **********************************
#> * TAM 3.3-4 (2019-07-17 18:34:22)
packageVersion("TAM")
#> [1] '3.3.4'
# from ?tam.se
#############################################################################
# EXAMPLE 2: Standard errors differential item functioning
#############################################################################
data(data.ex08)
formulaA <- ~ item*female
resp <- data.ex08[["resp"]]
facets <- as.data.frame( data.ex08[["facets"]] )
# investigate DIF
mod <- TAM::tam.mml.mfr( resp=resp, facets=facets, formulaA=formulaA, verbose = FALSE)
#> --- Created person identifiers.
# summary(mod)
# estimate standard errors
semod <- TAM::tam.se(mod)
#> Item parameters
#> |**********|
#> |----------|
#> Regression parameters
#> |*|
#> ||
#> |-|
mod2 <- TAM::tam.mml.mfr( resp=resp, facets=facets, formulaA=formulaA, verbose = FALSE,
control = list(snodes = 2000))
#> --- Created person identifiers.
# summary(mod)
# estimate standard errors
semod2 <- TAM::tam.se(mod2)
#> Item parameters
#> |**********|
#> |
#> Error in matrix(thetasamp.density, nrow = nstud, ncol = ncol(gwt), byrow = TRUE): 'data' must be of a vector type, was 'NULL'
Dear Alexander, I have problems with standard error estimation via
tam.se()
for models that were estimated using QMC, see example below. I don't know if this is a feature or a bug. But in case it's a bug, I propose the committed changes. I tried a few examples and compared the SEs for models with and without QMC, and they were comparable IMHO. Feel free to adapt this to your needs. Best, HansjörgCreated on 2019-07-19 by the reprex package (v0.3.0)