Description Usage Arguments Details Value References See Also Examples
Simulation for factor copulas for ordinal item response data, all bivariate linking copulas in same parametric family for each factor
1 2 | sim1irfact(ucuts,n,parobj1,qcond1,copname1,ivect=F)
sim2irfact(ucuts,n,parobj1,parobj2,qcond1,qcond2,copname1="",copname2="",ivect=F)
|
ucuts |
(ncat-1)xd matrix of cutpoints, increasing in each column and bounded in (0,1); ncat=number of categories |
n |
sample size |
parobj1 |
parameter vector of dimension d or parameter matrix with d rows for factor 1, where d is dimension of factor copula; this is dx2 for something like BB1 copula |
parobj2 |
parameter vector of dimension d or parameter matrix with d rows for factor 2 |
qcond1 |
function for copula conditional inverse cdf C_{U|V}^{-1}(u|v), choices include qcondfrk, qcondgum, qcondgumr, qcondbb1, qcondbvtcop with fixed nu1. |
qcond2 |
function for copula conditional inverse cdf C_{U|V}^{-1}(u|v) for second factor, choices include qcondfrk, qcondgum, qcondgumr, qcondbvtcop with fixed nu2. |
copname1 |
copula name: the function checks on "frank", "mtcj", "mtcjr", "fgm" for which qcond has closed form. |
copname2 |
copula name for factor 2 |
ivect |
flag that is T if qcond1 and qcond2 have vectorized forms |
These interface to sim1fact and sim1fact, with discretization based on ucuts.
data matrix of dimension nxd with values in 0,...,ncat-1
Nikoloulopoulos A K and Joe H (2015). Factor copula models for item response data, Psychometrika, 80, 126-150.
1 2 3 4 5 6 7 8 9 10 |
ucuts=matrix(c(.3,.6,.4,.7,.5,.8),2,3)
param=c(5.5,6.5,4)
set.seed(123)
ydat=sim1irfact(ucuts,n=1000,param,qcond1=qcondfrk,copname1="frank")
print(cor(ydat))
for(j in 1:length(param)) print(table(ydat[,j]))
ydat2=sim2irfact(ucuts,n=1000,param,c(2,2,2),qcond1=qcondgum,qcond2=qcondgum,ivect=TRUE)
print(cor(ydat2))
for(j in 1:length(param)) print(table(ydat2[,j]))
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.