I have a dataset consisting of 11 items of binary data (0/1) that will be the basis for my dependent variable. There are 240 observations collected by means of a survey. Since the data is binary, the CFA is run based on summary statistics (SSD) and using polychoric correlations. See code below (as also previously supplied in this forum):
----
local thevars ..."list of variables goes here"...
polychoric `thevars'
mat polychR = r(R)
forvalues i=1/`: word count `thevars' ' {
forvalues j=1/`i' {
local setcor `setcor' `=polychR[`i',`j']'
}
if `i' < `: word count `thevars' ' local setcor `setcor' \
}
local N = _N
clear
ssd init `thevars'
ssd set obs `N'
ssd set cor `setcor'
..."sem or factormat with options start here..."
----

This code has provided me with a good model, the problems I have is with figuring out how to be able to predict factor-scores for my observations. As the Stata-memory only holds summary statistics after the sem-command, I can not run “predict” as I could have if based on the raw data. How should I proceed in order to predict factor-scores for my observations?