Variable Selection and Model Averaging in Semiparametric Overdispersed Generalized Linear Models - Statistics > MethodologyReport as inadecuate




Variable Selection and Model Averaging in Semiparametric Overdispersed Generalized Linear Models - Statistics > Methodology - Download this document for free, or read online. Document in PDF available to download.

Abstract: We express the mean and variance terms in a double exponential regressionmodel as additive functions of the predictors and use Bayesian variableselection to determine which predictors enter the model, and whether they enterlinearly or flexibly. When the variance term is null we obtain a generalizedadditive model, which becomes a generalized linear model if the predictorsenter the mean linearly. The model is estimated using Markov chain Monte Carlosimulation and the methodology is illustrated using real and simulated datasets.



Author: Remy Cottet, Robert Kohn, David Nott

Source: https://arxiv.org/







Related documents