I tried the shared frailty option in a cox proportional hazard model to account for the within-family correlation of my pedigree data.
After several hours (!!!) I got a convergence error when estimating the frailty variance.
This just happened in a larger dataset (for example in 1500 individuals from 600 families, 1-4 members within a family); in a smaller subset it's working fine. Is it a memory problem ? Are there any requirements for the coding of the shared frailty variable ?
I would appreciate any suggestions very much since I have to analyze this (total) dataset very soon.
Thanks much,
Marcus
I believe that the standard approach is to use a share
frailty model. This can be done using -stcox-'s
-share()- option.
Best,
Ricardo.
--- "Fischer, Marcus" <[email protected]> wrote:
> Hi,
>
> I wish to evaluate associations between prospective
> disease risk and one or more genetic or
> environmental covariates, using data from families
> ascertained because of the disease statuses of some
> of their members. The population consists of
> nonoverlapping families each containing one or more
> individuals. Both disease occurences and covariates
> may be correlated within families.
> Is it possible to account for the within-family
> correlation in a cox proportional hazard model ? If
> so, how ?
> Can I just use the cluster(family-identifier) option
> to work with the robust variance-covariance matrix ?
>
> Thanks for your help.
> Marcus
*
* 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/
