Generic Conditions for Forecast Dominance

TL;DR

This paper provides a theoretical framework for understanding when one forecast method outperforms another, especially in complex scenarios with uncalibrated or misspecified models, with applications in finance and economics.

Contribution

It introduces a new characterization of forecast dominance for the mean functional that accounts for non-nested information sets and uncalibrated forecasts, extending prior empirical analyses.

Findings

Derived a new theoretical characterization of forecast dominance.

Identified scenarios where dominance occurs despite model misspecification.

Validated relevance with empirical data from finance and economics.

Abstract

Recent studies have analyzed whether one forecast method dominates another under a class of consistent scoring functions. While the existing literature focuses on empirical tests of forecast dominance, little is known about the theoretical conditions under which one forecast dominates another. To address this question, we derive a new characterization of dominance among forecasts of the mean functional. We present various scenarios under which dominance occurs. Unlike existing results, our results allow for the case that the forecasts' underlying information sets are not nested, and allow for uncalibrated forecasts that suffer, e.g., from model misspecification or parameter estimation error. We illustrate the empirical relevance of our results via data examples from finance and economics.