Do Composite Procedures Really Improve the Accuracy of Outlook Forecasts?

This paper investigates whether the accuracy of outlook hog price forecasts can be improvedusing composite forecasts in an out-of-sample context. Price forecasts from four wellrecognizedoutlook programs are combined with futures-based forecasts, ARIMA, andunrestricted Vector Autoregressive (VAR) models. Quarterly data are available from 1975.Ithrough 2007.IV, which allow for a relatively long out-of-sample evaluation period afterpermitting model specification and appropriate composite-weight training periods. Results showthat futures and numerous composite procedures outperform outlook forecasts. At intermediatehorizons, OLS composite procedures perform rather well. The superiority of futures andcomposite forecasts decreases at longer horizons except for an equal-weighted approach.Importantly, with just few exceptions, nothing outperforms the equal-weight approachsignificantly in any program or horizon. Overall, findings favor the usage of equal-weightedcomposites, a result that is consistent with previous empirical findings and recent theoreticalpapers.