st: xtreg and xtlogit with time-invariant variables

[Shehzad Ali <drshehzad_ali@yahoo.com>]

Hi listers,

I am wondering if someone could advise on this. 

When xtreg and xtlogit models are implemented with fixed effects option, all 
time-invariant variables are dropped from the analysis. I am working with a 
dataset with response variable observed at 3 time points. The explanatory 
variable includes timepoint_var (obviously varies with time) and the others are 
time-invariant covariates that are dropped in the analysis. Could someone advise 
on the best way to model the relationship between response at multiple 
timepoints and the time-invariant variables? I should mention that the model 
runs when random effects option is specified.

Any help would be greatly appreciated.

Many thanks,
Shehzad



      

*
*   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/