Re: st: graphing Kaplan Meier confidence interval band

[Phil Clayton <philclayton@internode.on.net>]

I may be on the wrong track but have you tried:
sts gen km=s
sts gen ub=ub(s)
sts gen lb=lb(s)

Then something along the lines:
twoway (rarea ub lb _t, sort) (line km _t, sort)

Phil

On 25/02/2011, at 1:47 AM, Nicolás Rojas wrote:

> Dear STATA users,
>                                I am estimating KM surviving
> probabilities of staying in teaching. My data comes from a multi stage
> sample so I need to use weights.
> KM estimation with confidence intervals is easy with sts commands.
> 
> sts timevar, failure(event)
> sts graph, ci
> 
> Very nice confidence intervals are plotted with ci option. However the
> command does not support weighting. It is not possible to do:
> 
> sts timevar [pw=weights], failure(event)
> sts graph, ci
> 
> Does anybody know how to do that manually??? Plotting confidence
> intervals alone is easy if you use sts list. Howoever, I cannot graph
> the colored bands.
> 
> 
> Thank you, best regards,
> 
> Nico Rojas
> Research Assistant
> PUC Chile
> 
> *
> *   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/


*
*   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/