2025/06/05 by Moala, Fernando Antonio, Moala, Adriano Buran, Jerez Lillo, Nixon +1
#Bayesian inference #Gumbel distribution #objective priors #reference prior
paper · doi:10.57805/revstat.vi.648
The selection of priors is a critical aspect of Bayesian analysis, although the literature lacks studies concerning the application of the Gumbel distribution using different objective priors. We derive objective priors for the two-parameter Gumbel distribution and present a fully Bayesian analysis. Our primary goal is to choose a prior that represents a state of "little knowledge" a priori for both parameters. To yield this, we implement Markov Chain Monte Carlo algorithms to sample from the posterior distribution and to calculate the Bayes estimators. This investigation is made in the context of extreme weather events, using maximum rainfall data.