Rejoinder to the discussants of the two JASA articles `Frequentist Model Averaging' and `The Focused Information Criterion', by Nils Lid Hjort and Gerda Claeskens

TL;DR

This paper is a detailed response to expert discussions on two articles about model averaging and the focused information criterion, addressing major issues and themes raised by discussants.

Contribution

It provides a comprehensive rejoinder to critiques and comments on the authors' previous work on model averaging and the focused information criterion.

Findings

Clarifies the authors' perspectives on key issues

Addresses major themes raised by discussants

Highlights the importance of the discussed methodologies

Abstract

We are honoured to have our work read and discussed at such a thorough level by several experts. Words of appreciation and encouragement are gratefully received, while the many supplementary comments, thoughtful reminders, new perspectives and additional themes raised are warmly welcomed and deeply appreciated. Our thanks go also to JASA Editor Francisco Samaniego and his editorial helpers for organising this discussion. Space does not allow us answering all of the many worthwhile points raised by our discussants, but in the following we make an attempt to respond to what we perceive of as being the major issues. Our responses are organised by themes rather than by discussants. We shall refer to our two articles as `the FMA paper' (Hjort and Claeskens) and `the FIC paper' (Claeskens and Hjort).