Log-binomial regression

Mtani Njegere

unread,

Apr 12, 2021, 9:20:51 AM4/12/21

to R-inla discussion group

Hello INLA team,

I have my dataset which is hierarchical multilevel with binary outcome variable. I want to evaluate the relative risk in each category of explanatory variable using log-binomial regression. How can fit this in INLA. The proposed R codes will be very helpful.

Helpdesk

unread,

Apr 12, 2021, 1:46:09 PM4/12/21

to Mtani Njegere, R-inla discussion group

can you write down the math for what model you want to fit, or make a R-
script simulating from the model?

> --
> 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/c6c7475f-ab61-4536-978f-5355d893c02an%40googlegroups.com
> .

--
Håvard Rue
he...@r-inla.org

Prince Michael Amegbor

unread,

Apr 12, 2021, 2:30:03 PM4/12/21

to R-inla discussion group

Hi,

This is my suggestion (I could be wrong). In normal glm you can specify log-binomial as

glm(formula, data=data, family=binomial(link="log"))

hence my deduction is that in INLA you can do

inla(formula, family='binomial',Ntrials="", data=data, control.family=list(link='log') ...)

I stand corrected. And as Håvard said you may get the appropriate answer by writing the math of your model.

Regards,

Reply all

Reply to author

Forward