Message-ID: <de1857f2-8a31-5923-0818-9174e58b2cd0@gmail.com>
Date: 2019-05-09T17:33:17Z
From: Ben Bolker
Subject: Extract individual likelihood from glmmTMB
In-Reply-To: <CANZfssyPua39rU7QuXe-a5rVtp76GgrqrtbryOiJQ9wg0DLRKQ@mail.gmail.com>
By "individual likelihood" do you mean you want the log-likelihood of
a particular model? logLik(fitted_model) should do it. (This is a
generic method that should work for most model types in R, although you
do have to be a bit careful in comparing log-likelihood values across
models from different packages as they may make different choices about
which constant terms to include/exclude.)
If you want something else, please clarify ...
cheers
Ben Bolker
On 2019-05-09 4:14 a.m., Suna Park wrote:
> Dear R-sig-mixed-models members,
>
> Hello, this is Suna.
>
> I'm studying the use of medical care for the elderly using zero-inflated
> Poisson and negative binomial mixed model with random intercept.
>
> And I'm considering how to extract the individual likelihood from the
> glmmTMB for Vuong(1989)'s test.
>
> Looking at the source code of the glmmTMB, it seems that the likelihood
> function is in the 'glmmTMB.dll'.
>
> But I can't open and see the function in the 'glmmTMB.dll'.
>
> Could you let me know about how to modify the likelihood function or
> extract individual likelihood?
>
> Thank you,
>
> Suna Park
>
>
> *Suna Park*
> College of Economics & Finance, Hanyang University,
>
> [[alternative HTML version deleted]]
>
> _______________________________________________
> R-sig-mixed-models at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>