[R-sig-ME] MCMC model selection reference

[emeline mourocq]

Dear Jarrod,

 

In your previous comment about DIC you said : "For non-Gaussian data I never
use DIC, and have seriously considered removing it from MCMCglmm." What is
then the alternative(s) you suggest to use to do model selection using
MCMCglmm for non-Gaussian data?

 

I am asking because I am investigating the effect of several predictors on a
binary one-inflated response variable using MCMCglmm and was planning to do
model selection based on delta DIC of model with and without the predictor
of interest.

 

Would you have a recommendation for "a best method" to use to do model
selection in my case?

 

Best regards

Emeline

 

 

 



---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus

	[[alternative HTML version deleted]]