Orthogonalized smoothing for rescaled spike and slab models

TL;DR

This paper introduces orthogonalized smoothing techniques for rescaled spike and slab Bayesian models, enhancing their application in high-dimensional linear regression and prediction problems involving orthogonal polynomials.

Contribution

It extends rescaled spike and slab models with orthogonalized smoothing, integrating local regression and generalized ridge regression for improved model selection and visualization.

Findings

Enhanced model selection in high-dimensional settings

Effective degrees of freedom for curvature visualization

Application to orthogonal polynomial prediction

Abstract

Rescaled spike and slab models are a new Bayesian variable selection method for linear regression models. In high dimensional orthogonal settings such models have been shown to possess optimal model selection properties. We review background theory and discuss applications of rescaled spike and slab models to prediction problems involving orthogonal polynomials. We first consider global smoothing and discuss potential weaknesses. Some of these deficiencies are remedied by using local regression. The local regression approach relies on an intimate connection between local weighted regression and weighted generalized ridge regression. An important implication is that one can trace the effective degrees of freedom of a curve as a way to visualize and classify curvature. Several motivating examples are presented.