Scale.model option in inla

[Alokesh Manna]

Hi INLA Members,

I have a question regarding scale.mode. I see that

scale.model

Logical. If TRUE then scale the RW1 and RW2 and BESAG and BYM and BESAG2 and RW2D models so the their (generlized) variance is 1. Default value is inla.getOption("scale.model.default")

Now, say if I use my model,

Y`X+f(ID,X,model="besag") 

In that case If I use scale.model=T, will it resolve identifiability issue? If not how can we do so? (as it is a spatially varying coefficient part included, I keep const=F)

P.S.: Identifiability issue comes in my opinion as the model seems to be Y=intercept+X \beta_1 + X \beta_ID,X = intercept+X(\beta1+\beta_ID,X)

Alokesh.

[Helpdesk (Haavard Rue)]

there is a vignette discussing this option in detail

> --
> 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/ac8f0a31-9f2c-494b-ab78-95950cd7fd5bn%40googlegroups.com
> .

--
Håvard Rue
he...@r-inla.org

[Alokesh Manna]

Hi Helpdesk,

Can you tell me what should I search for?

On Wed, 22 May 2024 at 1:55 PM, Helpdesk (Haavard Rue) <he...@r-inla.org> wrote:

*Message sent from a system outside of UConn.*

[Helpdesk (Haavard Rue)]

browseVigettes(package="INLA")

std R stuff...

--
Håvard Rue
he...@r-inla.org