[R-sig-ME] Convergence in binomial MCMCglmm

[David Duffy]

On Wed, 8 Jun 2011, Jarrod Hadfield wrote:

> I'm not sure that censoring can be adequately handled when treating
> survival as a series of 0 binary outcomes ending in a 1, but I would be
> interested if anyone knows how it could be dealt with?

I would have thought about this problem as a "conventional" mixed model 
for survival data with time varying covariates: so as a "counting process" 
model.  The coxme package can fit this - I have only used it for ordinary 
right censored type data, so I can't give specific advice.  To do this 
kind of thing in an MCMC framework, I think one usually does it as a 
piecewise exponential eg

Scurrah K et al Variance components analysis for pedigree-based censored 
survival data using generalized linear mixed models (GLMMs) and Gibbs 
sampling in BUGS. Genetic Epidemiology Volume 19, Issue 2, pages 127-148, 
September 2000.

You can ask Katrina S about this.


Cheers, David Duffy.

-- 
| David Duffy (MBBS PhD)                                         ,-_|\
| email: davidD at qimr.edu.au  ph: INT+61+7+3362-0217 fax: -0101  /     *
| Epidemiology Unit, Queensland Institute of Medical Research   \_,-._/
| 300 Herston Rd, Brisbane, Queensland 4029, Australia  GPG 4D0B994A v