The Practical Implementation of Bayesian Model Selection
Hugh Chipman, Edward I. George, Robert E. McCulloch · Lecture notes-monograph series · 2001
In principle, the Bayesian approach to model selection is straightforward. Prior probability distributions are used to describe the uncertainty surrounding all unknowns. After observing the data, the posterior distribution provides a coherent post data summary of the remaining uncertainty which is relevant for model selection. However, the practical implementation of this approach often requires carefully tailored priors and novel posterior calculation methods. In this article, we illustrate some of the fundamental practical issues that arise for two different model selection problems: the variable selection problem for the linear model and the CART model selection problem.
Cited by 414 works. doi:10.1214/lnms/1215540964
Authors
References
- Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing
- Classification and Regression Trees.
- Classification and Regression Trees.
- Stochastic Relaxation, Gibbs Distributions, and the Bayesian Restoration of Images
- Induction of Decision Trees
- Estimating the Dimension of a Model
- Bagging Predictors
- Information Theory and an Extension of the Maximum Likelihood Principle
- Monte Carlo sampling methods using Markov chains and their applications
- Ideal Spatial Adaptation by Wavelet Shrinkage