Other hyperpriors for INLA

[Alokesh Manna]

Hi INLA group,

Good afternoon!

I was wondering if we can use in besag the tau( NOT theta= log(tau) ) with

1. truncated normal
2. Half cauchy
3. Inverse gamma

Currently I see the hyperparameter specifications are the default loggamma with a hyperparameter (1,1e10^-5). Is it making from gamma and then transferring to loggamma? Why was it made internally for loggamma?

I think loggamma is giving some sort of convergence issues when you have approximately 10 random effects. 

[Helpdesk (Haavard Rue)]

you can check these for how to define your own priors.

note that the internal parameterisation is fixed, so you need to give the log
prior in terms of theta = log(precision), for example,

On Tue, 2024-11-19 at 16:39 -0800, Alokesh Manna wrote:
> Hi INLA group,
>
> Good afternoon!
>
> I was wondering if we can use in besag the tau( NOT theta= log(tau) ) with
>

>    1. truncated normal
>    2. Half cauchy
>    3. Inverse gamma

>
> Currently I see the hyperparameter specifications are the default loggamma
> with a hyperparameter (1,1e10^-5). Is it making from gamma and then
> transferring to loggamma? Why was it made internally for loggamma?
>
> I think loggamma is giving some sort of convergence issues when you have
> approximately 10 random effects. 
>
>

> --
> 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, visit
> https://groups.google.com/d/msgid/r-inla-discussion-group/21ad923e-7779-4362-ba29-0c5228819126n%40googlegroups.com
> .

--
Håvard Rue
he...@r-inla.org

[Helpdesk (Haavard Rue)]

> inla.doc("rprior")
> inla.doc("expression")
> inla.doc("table:")