search

jasp-stats / jasp-issues

This repository is solely meant for reporting of bugs, feature requests and other issues in JASP.
56 stars 29 forks source link

Bayesian Poisson regression #1439

Open TarandeepKang opened 2 years ago

TarandeepKang commented 2 years ago
In a partial repeat of #59 I would like to again request the addition of poisson regression models, in their various forms. Just today I came across this paper, which presents the Bayesian implementation, along with associated code: D’Angelo, L., & Canale, A. (2023). Efficient Posterior Sampling for Bayesian Poisson Regression. Journal of Computational and Graphical Statistics, 32(3), 917–926. https://doi.org/10.1080/10618600.2022.2123337 I am not knowledgeable enough to determine if this Bayesian implementation is the ideal one, but I trust one of the team will be able to determine this.
Kucharssim commented 2 years ago

Dear @TarandeepKang,

thank you for the suggestion! Support of GLMs including Poisson regression is something we have on our to-do list of features that we would very much like to offer; I cannot give you the exact timeline, but it's something we plan to work on very soon!

TarandeepKang commented 2 years ago

Hello again, I've just come across this new paper describing a new GLM method. The authors seem to suggest that it might be superior in some cases to existing methods, and I thought it might be worthy of your consideration, when you consider updating the GLM procedures!

Deen, M., & de Rooij, M. (2020). ClusterBootstrap: An R package for the analysis of hierarchical data using generalized linear models with the cluster bootstrap. Behavior research methods, 52(2), 572-590.

tomtomme commented 5 months ago

@TarandeepKang Frequentist Poisson is done for GLM and GLMM. Also for Bayesian GLMM. So I guess this leaves Bayesian GLM. Please update issue title to reflect this. Thx :)