Dear Statalist,
I apologize for maybe asking something real simple but I stuck on the
following problem:
I have a panel dataset in which unit i=1, .., N is oberserved for
t=1,.., n(i) times. The response y(it) for unit i at time t is binary
and there are some subject specific x(it) and time specific z(t)
covariates. In addition there are some unobserved time specific factors
which cause the units to be cross-sectional dependent. Now I want to
estimate the following model:
P(y(it) = 1|x(it-1),z(t-1)) = F(ß(0)+ßx(it-1)+bz(t-1)+v(t)),
where ß(0), ß and b are parameters and F is for example the logistic
function. I have two questions:
1) How can I estimate the time specific random effect v(t) in stata?
What I see in the documentation for xtlogit is the subject specific
random effect v(i). Must I adjust the index variables via the i() and
t() options in the xt command? (my time variable t is the cluster
variable i in stata?)
2) How can I incorporate the lagged covariates x(it-1) and z(t-1) into
the estimation?
Any suggestions are appreciated very much!
Thomas
*
* 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/