Discussion of "Fast Approximate Inference for Arbitrarily Large Semiparametric Regression Models via Message Passing"

TL;DR

This discussion paper reviews and comments on Wand's work on fast approximate inference methods for large semiparametric regression models using message passing algorithms.

Contribution

It provides insights and critical analysis of Wand's proposed inference technique, highlighting its strengths and potential limitations.

Findings

Highlights the efficiency of message passing in large-scale semiparametric models

Identifies potential areas for improvement in Wand's approach

Discusses the applicability of the method to various regression scenarios

Abstract

Discussion paper on "Fast Approximate Inference for Arbitrarily Large Semiparametric Regression Models via Message Passing" by Wand [arXiv:1602.07412].