RE: st: gllamm vs xtlogit results

["Anderson, Bradley J" <BAnderson1@lifespan.org>]

Wanted to thank Jean Marie, Austin Nichols, and Hind Sbihi for their helpful comments.  Since yesterday I've learned that the data I'm working with probably require a large number of integration points; the intraclass correlation is quite high and the average number of level 1 observations per group is 87.5.  Even models estimating only a constant and the random effect were not converging until a large number of integration points had been specified.  I also think I need to re-think model specification cause when I think about the data the assumption that the X are not associated with the random effect is not reasonable, especially for one of the predictors.  Thanks again for the very useful comments.

Brad


*
*   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/