Having issues understanding the "seasonal" model

[Sebastian Rodriguez]

Hello all,

I am having some issues understanding what the "seasonal" model is doing.

I have panel data with timeseries from over 800 locations. I want to add a seasonal structure to my data, as my data is the monthly incidence rate of a disease. I am using 

f(time, model = "seasonal", season.length = 12).

Is this term adding a seasonal structure to my response or to my error term?

Thanks in advance and sorry if my question is too general.

Best,

Sebastian

[Helpdesk]

with

y ~ 1 + x + f()

then with Gaussian likelihood, then ``1 + x + f()'' is the linear
predictor, and the likelihood account for the observational noise.

if you consider ``1 + x'' as the linear predictor, then -f(), plus the
iid for the Gaussian likelihood, will be the new likelihood, and -f()
goes into the observational `noise''.

> --
> 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/9d8d3a00-2981-450c-9584-5fb65fd6b51en%40googlegroups.com
> .

--
Håvard Rue
he...@r-inla.org

[Fredrik L. Aanes]

Is this equivalent to a random walk of order 1 or 2 with constraints X(1) = X(12), X(2) = X(11), X(3)= X(10), X(4) = X(9), X(5) = X(8), X(6) = X(7)? 

[Fredrik L. Aanes]

Aah. That would create a symmetric pattern! Cyclic perhaps?