Bayesian model selection for linear regression

TL;DR

This paper introduces a Bayesian approach to linear regression with basis functions, enabling automatic model selection that balances data fit and complexity to prevent overfitting.

Contribution

It presents a Bayesian model selection method for linear regression that incorporates Occam's razor to choose the optimal model complexity.

Findings

Effective automatic model selection using Bayesian criteria

Prevents overfitting by balancing model complexity and data fit

Applicable to linear regression with basis functions

Abstract

In this note we introduce linear regression with basis functions in order to apply Bayesian model selection. The goal is to incorporate Occam's razor as provided by Bayes analysis in order to automatically pick the model optimally able to explain the data without overfitting.