Dear Statalist,
I'm running a loglinear model with 4-way-interaction (all variables are
categorical). Up to now I have created the relevant dummies manually but
- as I have to specify several versions of the model - I was very happy
to find that there is also a possibility of creating them automatically
by the means of -desmat-. I used it as a prefix (one can use it as a
command, too) and then compared the output with the results of my
manually produced model.
The both results (including the algebraic sign of some effects) are
different! The log likelihood value, the Pearson Chisq and the AIC value
is identic but the deviance and BIC is higher in the desmat output than
in the manual model.
Can anybody tell me what the reason for it could be? Has anybody
experience with -desmat- ?
(The latest version of -desmat- is available at SSC-Ideas (reference:
http://www.xs4all.nl/~jhckx/desmat/stata/))
Is there any other command / prefix for automatic specifying
4-way-interactions?
Any help would be very appreciated!
Joan
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
