Tests for Forecast Encompassing

1998/04/01 by David I. Harvey, Stephen J. Leybourne, Paul Newbold · 2 citations
Decision Sciences · Economics, Econometrics and Finance · Mathematics · #Advanced Statistical Methods and Models #Forecasting Techniques and Applications #Monetary Policy and Economic Impact

paper · doi:10.1080/07350015.1998.10524759

openalex publication_date 1998/04/01 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/30

Abstract

We consider the situation in which two forecasts of the same variable are available. The possibility exists of forming a combined forecast as a weighted average of the individual ones and estimating the weights that should be optimally attached to each forecast. If the entire weight should optimally be associated with one forecast, that forecast is said to encompass the other. A natural test for forecast encompassing is based on least squares regression. We find, however, that the null distribution of this test statistic is not robust to nonnormality in the forecast errors. We discuss several alternative tests that are robust.

Citations

• Experience with Forecasting Univariate Time Series and the Combination of Forecasts
• The Combination of Forecasts
• Testing the equality of prediction mean squared errors
• A Heteroskedasticity-Consistent Covariance Matrix Estimator and a Direct Test for Heteroskedasticity
• Comparing Predictive Accuracy

Cited by

• Emerging stock market volatility and economic fundamentals: the importance of US uncertainty spillovers, financial and health crises
• Approximately normal tests for equal predictive accuracy in nested models