Time-varying Forecast Combination for High-Dimensional Data

TL;DR

This paper introduces a novel nonparametric, time-varying forecast combination method that adapts to structural changes in high-dimensional data, demonstrating superior performance in simulations and empirical studies.

Contribution

It develops a new estimator with oracle properties for high-dimensional forecast combination, handling structural changes effectively.

Findings

Outperforms existing methods in simulations with structural changes

Successfully selects relevant forecasts with high probability

Shows improved accuracy in inflation and equity premium predictions

Abstract

In this paper, we propose a new nonparametric estimator of time-varying forecast combination weights. When the number of individual forecasts is small, we study the asymptotic properties of the local linear estimator. When the number of candidate forecasts exceeds or diverges with the sample size, we consider penalized local linear estimation with the group SCAD penalty. We show that the estimator exhibits the oracle property and correctly selects relevant forecasts with probability approaching one. Simulations indicate that the proposed estimators outperform existing combination schemes when structural changes exist. Two empirical studies on inflation forecasting and equity premium prediction highlight the merits of our approach relative to other popular methods.