Firth‘s penalization for logistic regression for rare events

[FredrikLAa]

Hi,

Is it possible to obtain the Firth's penalized regression in INLA, where

the log likelihood is given by:

L(Beta)det(X'WX)^0.5,

where W=diag(pi_i(1-pi_i))

see e.g slide 10 in https://www.eio.upc.edu/ca/seminari/docs/georg-heinze-logistics-regression-with-rare-events-problems-and-solutions.pdf

I have a situation where there are very few "positive" cases, 320 out of 42 000

Any alternatives to Firth's penalized regression that are implemented in INLA?

Best

Fredrik

[Helpdesk (Haavard Rue)]

in short, no. should be easy to do this through some other packages
though

> --
> 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/af4c301e-4eff-4aed-8590-6236d96fc3cen%40googlegroups.com
> .

--
Håvard Rue
he...@r-inla.org