Gumbel type in GEV distribution

paul-buerkner / brms

brms R package for Bayesian generalized multivariate non-linear multilevel models using Stan

https://paul-buerkner.github.io/brms/

GNU General Public License v2.0

1.28k stars 187 forks source link

Gumbel type in GEV distribution #1426

Closed JackCaster closed 2 years ago

JackCaster commented 2 years ago

According to https://en.wikipedia.org/wiki/Generalized_extreme_value_distribution, there are three types of GEV: Gumbel, Frechet, and Weibull. By reading the vignette brms families, it looks like brms parametrizes the GEV by including only when xi is different from 0 (i.e., no Gumbel case). According to the Wikipedia article, when xi = 0 then t(x) = exp{-(x-mu)/sigma}.

Would you be able to confirm if GEV in brms includes all 3 types of distributions or only the Frechet / Weibull ones?

paul-buerkner commented 2 years ago

Have you tried setting xi=0 and see how it works? In the future, please ask brms related questions on https://discourse.mc-stan.org/

JackCaster commented 2 years ago

I apologize I did not use the forum. I set xi = 0 into dgen_extreme_value and it yields a result. I take it then that the Gumbel type is implemented, which is great. Thank you for the quick reply.