On Mon, 20 Oct 2003 09:54:03 -0400 "Paul E. Karner" <[email protected]>
wrote:
> Hello,
>
> I am trying to ascertain whether RHS variables are allowed to be
time-varying in certain distributions using the 'streg' command for a
survival-time dataset. The manual seems to say that with Weibull
regressions, if the 'time' option is specified (i.e. the model is run
in accelerated failure time form as opposed to proportional hazard),
STATA uses constant covariates. Are covariates also treated as
constant when the Gamma distribution (only implemented in STATA as
accelerated failure time) is specified, or is there a way to have
time-varying covariates in the Gamma models? >
> Perhaps an answer could be found if there is a way to view (portions
of) the X matrix of regressors that were used in the model - is this
possible? >
You can have explanatory variables that vary with time in -streg-
models, but you need to episode-split your data appropriately first.
See [ST] manual entry on -stplit-. [You split observed survival times
into sub-periods of your choosing, and it is within those subperiods
that covariates are assumed to take constant values.]
Stephen
----------------------
Professor Stephen P. Jenkins <[email protected]>
Institute for Social and Economic Research (ISER)
University of Essex, Colchester, CO4 3SQ, UK
Tel: +44 (0)1206 873374. Fax: +44 (0)1206 873151.
http://www.iser.essex.ac.uk
*
* 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/
