Multiple adjacency matrix for spatiotemporal model

[MUHAMMED TAHIR MUHAMMED]

Hello colleagues,

I am trying to model cases of malaria in Nigeria over time using adaptive spatiotemporal modeling approach. However, I am having multiple adjacency matrices of which each year (2010, 2015, 2018 and 2021) have its own unique adjacency matrix, W_t.

In my effort to adapt this model, I specify my model as:

 formula <- Y ~ 1 + covariate1 + covariate2 +

f(region, model = "bym2", graph = W1) + 
f(region, model = "bym2", graph = W2) + 
f(region, model = "bym2", graph = W3) + 

f(region, model = "bym2", graph = W4) + 
f(time_periods, model = "ar1").

Is this specification correct?

I have searched many literatures on adaptive spatiotemporal models but could not find this particular scenario.  PLEASE I NEED YOUR CONTRIBUTIONS AND SUGGESTIONS TO THE PROGRESS OF THIS WORK.

Thanks.       

[Helpdesk (Haavard Rue)]

you need to do

region1 <- region
region2 <- region
region3 <- region
region4 <- region
region5 <- region


and the use

...+ f(region1, model=...) + f(region2, model=...) +
f(region3, model....)

further: I would guess region1 should be is NA except in 2010, region2 is NA
except in 2021, etc so you would need to change this.

technically, this is an instance of regional data or aggregated data, for which
the 'correct' approach is to have a continous space-time latent field, which is
then observed diffrently for each year. just keep that in mind

> --
> 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/95e44f1a-a67a-4493-8a79-0e176f2b1517n%40googlegroups.com
> .

--
Håvard Rue
he...@r-inla.org