Re: st: selection in continuous time survival framework without instruments
Date
Thu, 29 Aug 2002 10:05:42 -0400
Alex Cowell asked:
1. Does anyone know of a way of using an individual fixed-effects (in
economics lingo) that would help handle this bias in a continuous time
framework?
2. Any other ideas on controlling for this kind of unobserved heterogeneity
bias? Again, remember there are no obvious identifying instruments. And
note that I think this is subtly different from modeling 'frailty' (could be
wrong on that though).
-----------------
This is something I've been interested in, too. Take a look at Hamilton &
Hamilton (1997) "Estimating surgical volume - outcome relationships
applying survival models: accounting for frailty and hospital fixed
effects" Health Economics v. 6:383-395, which is based on the work of
Heckman and Singer (1984) "A method for minimizing the impact of
distributional assumptions in econometric models for duration data"
Econometrica v. 62:271-320.
I could never get this programmed into Stata. (Hamilton and Hamilton's work
was done in .tsp) Frailty is, I believe, identical to unobserved
heterogeneity, and there are some frailty survival models canned in Stata
that might work for you, if you want to work in a discrete-time framework
(which it sounds like you don't).
Good luck - wish I could be more helpful,
Heather Gold
