Message-ID: <Pine.A41.4.61b.0505311103500.190182@homer05.u.washington.edu>
Date: 2005-05-31T18:07:59Z
From: Thomas Lumley
Subject: Shared Frailty in survival package (left truncation, time-dep. covariates)
In-Reply-To: <005301c56607$d77217b0$1b9e5f86@simurechner>
On Tue, 31 May 2005, Stefan Pohl wrote:
> Dear list,
>
> I want o fit a shared gamma frailty model with the frailty specification
> in the survival package.
>
> I have partly left-truncated data and time-dependent covariates. Is it
> possible to combine these two things in the frailty function. Or are the
> results wrong if I use data in the start-stop-formulation which account
> for delayed entry?
It should work: as usual in the Cox model you get left-truncation and
time-dependent covariates for free, since the partial likelihood only
considers one risk set at a time. You do need to make sure you specify
the correct id variable, of course, and left truncation may reduce the
amount of information available for estimating the frailty.
-thomas