st: Predict after Factormat

["David Doyle" <david.doyle@dcu.ie>]

Dear Stata List Users,

I am trying to run a factor analysis with missing data. Currently, I am
following the procedure for factor analysis with missing data outlined on
the IDRE website at UCLA, using maximum likelihood with the
expectation-maximization (EM) algorithm to estimate the factor analysis from
the covariance matrix:
http://www.ats.ucla.edu/stat/stata/faq/factor_missing.htm. 

This works fine, but I am having trouble using ?predict? to generate a new
variable after the factor analysis. The help files state I need to use the
means() option after factormat, but I cannot come across an example of the
syntax, and lots of different attempts I have made have been unsuccessful.

Does anyone know what I need to do to run predict after factormat? What is
the correct command?

The commands in the do-file I am using are as follows:

******
count /*count total number of observations */
 
mi set mlong
 
mi register imputed varlis1-33
   
mi impute mvn varlist1-33, emonly

matrix cov_em = r(Sigma_em)

matrix list cov_em

factormat cov_em, n(*) fact(1) pcf

rotate, varimax

****

I wish to follow this with predict, but currently I get his message: 

(variable means assumed 0; use means() option of factormat for nonzero
means)

Any help would be greatly appreciated.

Best,

Dave. 


Dr. David Doyle
Lecturer in Politics
Director of the PhD Programme
School of Law & Government
Dublin City University



*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/