A0383
Title: An extreme value Bayesian Lasso for the conditional bulk and tail
Authors: Patricia de Zea Bermudez - FCiencias.ID - Associacao para a Investigacao e Desenvolvimento de Ciencias (Portugal)
Miguel de Carvalho - FCiencias.ID - Associacao para a Investigacao e Desenvolvimento de Ciencias (Portugal) [presenting]
Abstract: A novel regression model for the conditional bulk and conditional tail of a possibly heavy-tailed response is introduced. The proposed model can be used to learn the effect of covariates on an extreme value setting via a Lasso-type specification based on a Lagrangian restriction. The model can be used to track if some covariates are significant for the bulk, but not for the tail---and vice-versa; in addition to this, the proposed model avoids the need for conditional threshold selection in an extreme value theory framework. The finite-sample performance of the proposed methods is assessed by means of a simulation study that shows that our method recovers the true conditional distribution over a variety of simulation scenarios, along with being accurate in variable selection. Rainfall data are used to display how the proposed method can learn to distinguish between key drivers of moderate rainfall, against those of extreme rainfall.
