Comparing the spatial effects of two models

[Prince Michael Amegbor]

hello,

I have a question about comparing the spatial effects of two models using r-inla. Assuming I have two models: Model 1 and Model 2. Where Model  1 is a survival model with the formula (adopted from Martine, Akerkar & Rue (2011):

f1 = inla.surv(time, event)∼ PM+ f(ID, model="iid", param=c(1, 1))

and Model 2 is expressed as 

f1 = inla.surv(time, event)∼ PM+ age + edu + inc + f(ID, model="iid", param=c(1, 1))

where ID is an index for spatial units.

Is there a way to compare the spatial effects from the two models in terms of change in the spatial risk? Can I just do such a comparison by looking at the differences in the log hazards for ID in the two models (e.g. Model 1 - Model 2)?

Thank you.

[Helpdesk]

Maybe you want to compare DIC or WAIC, if you're up to a model
comparison. Otherwise, if you're after the chance in the risk, you have
to check how to risk for the two fits differ.

> --
> You received this message because you are subscribed to the Google
> Groups "R-inla discussion group" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to r-inla-discussion...@googlegroups.com.
> To view this discussion on the web, visit
> https://groups.google.com/d/msgid/r-inla-discussion-group/787d7722-c56f-4794-be76-8b4033e784ccn%40googlegroups.com
> .

--
Håvard Rue
he...@r-inla.org