MCMC with negative binomial distribution

[Jacqueline]

Dear runmlwin users,

I am trying to analyse count data using a negative binomial distribution with a log link function. I have chosen a negative binomial distribution because the data exhibit overdispersion. As recommended for discrete response models, I am estimating the model with MCMC. When I run this specification for a relatively simple null model with two levels, I receive the message 'assertion is false'. However, the model runs fine when I estimate it using IGLS or if I change the distribution to Poisson. Could this message mean that a negative binomial distribution is not appropriate for the data?

As always, any help would be greatly appreciated!

Many thanks,

Jacqueline

[ChrisCharlton]

The negative binomial distribution is currently not implemented in MLwiN when using MCMC estimation. We will try to provide a better error message to indicate this in a later release of runmlwin.

[Jacqueline]

Hi Chris,

Thank you very much for your response.

Kind regards,

Jacqueline