Paper Image

A simplified Bayesian model for functional regression

Published on:

21 December 2023

Primary Category:


Paper Authors:

José R. Berrendero,

Antonio Coín,

Antonio Cuevas


Key Details

Proposes Bayesian methodology for functional regression using RKHS theory

Uses finite-dimensional approximation with parametric form based on kernel

Derives posterior consistency result based on classic theorem by Doob

Prediction strategies are competitive against common alternatives

Includes Bayesian-motivated variable selection procedure

AI generated summary

A simplified Bayesian model for functional regression

This paper proposes a Bayesian approach to functional regression based on reproducing kernel Hilbert spaces. It uses a finite-dimensional approximation with a parametric form to simplify inference. Theoretical results guarantee posterior consistency. Prediction strategies from the Bayesian model are competitive on simulations and real data.

Answers from this paper


No comments yet, be the first to start the conversation...

Sign up to comment on this paper

Sign Up