Fit statistics for binomial models

Welcome to the forum for R2MLwiN users. Feel free to post your question about R2MLwiN here. The Centre for Multilevel Modelling take no responsibility for the accuracy of these posts, we are unable to monitor them closely. Do go ahead and post your question and thank you in advance if you find the time to post any answers!

Go to R2MLwiN: Running MLwiN from within R >> http://www.bris.ac.uk/cmm/software/r2mlwin/
Post Reply
timcadman
Posts: 2
Joined: Wed Oct 16, 2019 1:16 pm

Fit statistics for binomial models

Post by timcadman » Wed Oct 16, 2019 1:19 pm

Hi there,

I'm using R2Mlwin to fit multilevel models using a binomial distribution. However neither the log-likelihood or deviance statistics are produced using this method. Are there any alternative statistics it is possible to extract from the model as an indicator of fit?

many thanks

ChrisCharlton
Posts: 1112
Joined: Mon Oct 19, 2009 10:34 am

Re: Fit statistics for binomial models

Post by ChrisCharlton » Wed Oct 16, 2019 2:31 pm

A Wald test will often be used for these kind of model (see chapter 9 of the MLwiN manual examples on https://www.bristol.ac.uk/cmm/software/ ... /examples/).

As MLwiN fits discrete models in (R)IGLS using quasi-likelihood it is usually advised to fit them using MCMC, at least for the final models. In this case you would have the DIC statistic available to use as an indicator for model fit and comparison.

timcadman
Posts: 2
Joined: Wed Oct 16, 2019 1:16 pm

Re: Fit statistics for binomial models

Post by timcadman » Mon Nov 04, 2019 10:48 am

Many thanks Chris

best,

Tim

Post Reply