The prior on the negative binomial overdispersion parameter can be a bit dicey. Take a look at this. The penalized complexity (PC) prior has worked well for me. There’s an implementation in this gist.
Related topics
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| Hierarchical Binomial Model - Not converging | 2 | 906 | February 1, 2019 | |
| Hierarchical Linear Regression Model -- why is it not converging? | 3 | 1586 | March 14, 2018 | |
| Hierarchical Binomial regression - ineffective sampling | 4 | 922 | January 17, 2018 | |
| Problem with convergence in hierarchical partially pooled model | 5 | 604 | October 19, 2018 | |
| Improving model convergence and sampling speed | 8 | 6342 | September 27, 2019 |