Please see the reference cited. If it's elusive, Google for Frank
Harrell's website to see summary statements.
Nick
[email protected]
Deepa Aggarwal
Thanks for quick response. But why is it bad?
On Thu, May 7, 2009 at 2:17 PM, Nick Cox <[email protected]> wrote:
> The usual term is Cox regression.
>
> You can use -stepwise- with -stcox-. It's a very bad idea, but it is
> possible. See Frank Harrell, Regression modeling strategies, Springer,
> NY, 2001.
>
> Nick
> [email protected]
>
> Deepa Aggarwal
>
> I am using stcox command of Stata for the longitudinal analysis in
time
> varying cox regression. Is there any command with which I can do
> backward
> or forward selection of variables in cox regression?
*
* 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/
