st: gllamm

[melaku.fekadu@gmail.com]

Thanks for Christopher and Richard for answering my previous question.

I have another question. I want to use -gllamm- to estimate a survival model and compare results from other models like --xtcloglog- and -pgmhaz-. I tried also -hshaz- but it takes a lot of time until convergence.

The problem with -gllamm- is that it even does not start iteration. It just says that 'variables have been dropped, can't continue'. 

I did _rmcoll and found only one variable which was collinear with other variable. i threw it out but it still keeps on telling me the same response over and over again, even as I threw out some more suspicious variables. 

I don't use any interaction variables in my model.

Does anybody have any idea how to tackle this problem?  

Thanks,
Melaku
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/