Coding repeated measures negative binomial distribution

Hi I'm a completely new user for R. I'm at the end of my thesis and there is just one data set that I need to analyse with R due to its nature, distribution etc. I normally use SPSS but in this instance it is not appropriate.

It was going to be a repeated measures ANOVA (two conditions (wild type mice and genetic knock outs), three days tested, 1 frequency measurement), with count data but a vast majority of the counts are 0. It was suggested I do a negative binomial (or Poisson) model to account for the data.

I'm in a situation where I ideally don't want to have to spend a large period of time learning all there is to know about R (at least until I graduate!), but just get some help with the repeated measures code.

Is there anyone that can help with this? I've attempted to install the package that I might need like gee, GLMM. Happy to provide more information and promise I will learn R when my PhD is over!

What I understand is that you want to apply ANOVA, but instead of assuming normality of the residuals, you'll be considering Negative Binomial distribution?

You know that ANOVA can be considered as Likelihood Ratio Test, right? So, you can do from that approach.

I've never used it, but possibly MASS::anova.negbin will be useful.

It'll use the output from MASS::glm.nb, and this page has a detailed example.

Hope this helps.

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.