Many thanks for your help. I will use the faq on first occurrences in panels to modify the if treat==1 statement accordingly.
At 11.36 06/04/2007 -0400, you wrote:
>Dear Nicola,
>
>You will find answers to your questions about inverse probability of treatment weighting (IPTW) (both theoretical and practical) in the writings of Jamie Robins and Miguel Hernan (both Harvard School of Public Health, they have a website with all the relevant papers).
>
>To answer your specific questions:
>
>(1) If treatment is marriage and you assume that a person remains married once they have been married, then the weight should be time-varying until (and including) the time of marriage, and exactly 1 for all observations thereafter. Reason: if you know that a person is married at time T than the probability that she is married in the following wave is exactly 1.
>
>(2) Your weights are time-varying. Stata's survival routines don't accept time-varying weights. Therefore, you'll have to use a discrete approximation of a survival model using logit or clogit (same thing if coded correctly, your choice).
>
>Hope this helps.
>
>Best,
>Felix
>
>
>
>Date: Wed, 04 Apr 2007 10:54:56 +0200
>From: <mailto:[email protected]>[email protected]
>Subject: Re: st: inverse probability of treatment weights
>
>Many thanks to both: the results are exactly the same in more than 95% of the cases, and their ratio is between .99999952 and .99999994 in the remaining ones. All ipw>1, while I expected some of them to be = 1. Remember that my data are panel and that my treatment variable is like "being married" in a world where no death/divorce is observable. Therefore e.g.
>id=1 t=1 : single (treatment=0)
>id=1 t=2 : marriage (treatment=1)
>id=1 t=3 : still married (treatment=1)
>...
>id=1 t=n : still married (treatment=1)
>Is it ok to have ipw > 1 for all n observations pertaining to id number 1, or should I manually correct it to 1 for the "still married" observations?
>
>What do you think about using some -st- command in lieu of the logit? When I was younger, I have been told that -st- are continuous time models, while logit is for discrete time ones. But then, I found several empirical papers using survival analysis with time-varying covariates measured once a year, as in my dataset. Is the choice between the two commands just a matter of taste??
*
* 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/
