I heard that adaptive quadrature has been suggested as a substitute of the quadrature estimation for "xtlogit". I am using Stata's "xtlogit", random effects command for two level generalized linear mixed models, which employ Gauss-Hermite quadrature to evaluate the log likelihood. However, quadrature performs poorly for some of the models.
Specifically, what command should I use to perform adaptive quadrature for cross-section time series data, with dichotomous dependent variable? Any recommendations for other alternative commands? Any references I should read?
Thanks a lot for your time!
Xiaoli
*
* 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/
