Re: st: stcrreg with categorical variables

[Maarten buis <maartenbuis@yahoo.co.uk>]

--- On Fri, 11/6/10, Sanam P wrote:
> I was wondering whether stcrreg (in stata11) can be used
> with categorical covariates with more than 2 levels, using
> below code, like how it worked with stcox command?

Yes

> I used above code however when I closed stata about half
> an hour after running the code, it was still in the process
> and hadn't finished the running yet! I am not sure what is
> the problem, could it be because that I am working with a
> big data set (around 30,000) ?

Big is in the eye of the beholder, I used to working with
datasets with a 100,000 observations, so I would not consider
that big, and there are others who would consider my datasets
small. 

Anyhow, I would first look at a tabulation of your categorical 
variables, and see if there are some categories that are 
exceptionaly small, as these tend to give you convergence 
problems. Also look for categories that all end up in one 
outcome or are all censored.

> Is there any way that I can find out how long is the
> estimated time left untill the running is finished?

Not really, with some experience you tend to be able to spot 
problematic itteration logs while they occur, but otherwise
no.

Hope this helps,
Maarten

--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany

http://www.maartenbuis.nl
--------------------------


      

*
*   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/