Thanks a lot Bobby for your useful suggestions. I have used the second
specification
stcox varlists, shared(country) efron nohr basehc(hazard)
Unfortunately I have another question for the list.
Using the command
stcurve, hazard kernel(gauss)
I achieve the plot of the smoothed hazard contribution for the analysis
time. I have 28 subjects (countries) and multiple-failures. Time is set
as time from origin.
May I plot the baseline hazard for each spell? Or is it another specification?
I mean, I?d like to analyze the time dependence, that is, the length of
the time already spent on the tranquil state as determinant of the probability
of exit into a failure.
Please any suggestion is welcome
Giorgio
__________________________________________________________________
Social price: l'ADSL diventa per tutti
Tiscali ADSL Senza Canone 640Kbps: GRATIS fino al 3 maggio costo
di adesione, attivazione e il modem per tutto il 2004. E per i primi 3 mesi,
con il Tiscali social price, navighi a 1,5 euro l'ora! Affrettati!
http://point.tiscali.it/adsl/prodotti/640Kbps/
*
* 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/
