Dear All,
I am trying to fit an stcox model to explain school drop out. I am
confused about the time varying covariates. I have time-varying
covariates in principle as I want to see how previous semester’s grade
affects this semester drop out for several semesters. However, it
makes little sense to me to use -tvc- for this. I was planning to
create some lagged variables separately and use them as a regular
covariates in the regression. Is there a more elegant way? Can a
combination of -tvc- and -texp(_t)- do the same thing? I also have a
set of time-invariant coefficients like gender, race, standardized
scores etc. Any help will be wonderful.
Regards,
Prabal
*
* 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/
