What is the default prior for "bym2" model?

[Shige Song]

Dear All,

I noticed that, with the newly unstable build of the INLA, it is now possible to estimate a "bym2" model without explicitly specifying priors. My question is: Is the default prior for BYM2 model PC prior?

Many thanks.

Best,
Shige

[Elias T. Krainski]

Yes it is. You can check with

      inla.models()$latent$bym2$hyper

Elias

--
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 post to this group, send email to r-inla-disc...@googlegroups.com.
Visit this group at https://groups.google.com/group/r-inla-discussion-group.
For more options, visit https://groups.google.com/d/optout.

[Håvard Rue]

Its the penalized complexity prior, that depends on the graph. It is
explained here,

@Article{art585,
author = {A. Riebler and S. H. S{\o}rbye and D. Simpson and H.
Rue},
title = {An intuitive {B}ayesian spatial model for disease
mapping that accounts for scaling},
journal = {Statistical Methods in Medical Research},
keywords = {year2016,my.own,my.cv},
year = 2016,
volume = {25},
number = {4},
pages = {1145-1165}
}




Recall, that you can also plot the priors in the same plot as the
posteriors, using

plot(result, plot.prior=TRUE)

On Thu, 2017-08-10 at 22:04 -0300, Elias T. Krainski wrote:
> Yes it is. You can check with
> inla.models()$latent$bym2$hyper
> Elias
>
> On 08/10/2017 05:16 PM, Shige Song wrote:
> > Dear All,
> >
> > I noticed that, with the newly unstable build of the INLA, it is
> > now possible to estimate a "bym2" model without explicitly
> > specifying priors. My question is: Is the default prior for BYM2
> > model PC prior?
> >
> > Many thanks.
> >
> > Best,
> > Shige
> > --
> > 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-group+unsubscribe@googlegroups.c
> > om.
> > To post to this group, send email to r-inla-discussion-group@google
> > groups.com.

> > Visit this group at https://groups.google.com/group/r-inla-discussi
> > on-group.
> > For more options, visit https://groups.google.com/d/optout.
>

> --
> 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 post to this group, send email to r-inla-discussion-group@googlegr
> oups.com.

> Visit this group at https://groups.google.com/group/r-inla-discussion
> -group.
> For more options, visit https://groups.google.com/d/optout.

--
Håvard Rue <hr...@r-inla.org>
r-inla.org