Conditionally Optimal Weights and Forward-Looking Approaches to Combining Forecasts

50 Pages Posted: 7 Apr 2017

See all articles by Christopher Gibbs

Christopher Gibbs

The University of Sydney - School of Economics

Andrey L. Vasnev

University of Sydney

Multiple version iconThere are 2 versions of this paper

Date Written: February 17, 2017

Abstract

In applied forecasting, there is a trade-off between in-sample fit and out-of-sample forecast accuracy. Parsimonious model specifications typically outperform richer model specifications. Consequently, there is often predictable information in forecast errors that is difficult to exploit. However, we show how this predictable information can be exploited in forecast combinations. In this case, optimal combination weights should minimize conditional mean squared error, or a conditional loss function, rather than the unconditional variance as in the commonly used framework of Bates and Granger (1969). We prove that our conditionally optimal weights lead to better forecast performance. The conditionally optimal weights support other forward-looking approaches to combining forecasts, where the forecast weights depend on the expected model performance. We show that forward-looking approaches can robustly outperform the random walk benchmark and many of the commonly used forecast combination strategies, including equal weights, in real-time out-of-sample forecasting exercises of inflation.

Keywords: Forecast Combination, Conditionally Optimal Weights, Forecast Combination Puzzle, Inflation, Phillips Curve

JEL Classification: C18, C53, E31

Suggested Citation

Gibbs, Christopher and Vasnev, Andrey L., Conditionally Optimal Weights and Forward-Looking Approaches to Combining Forecasts (February 17, 2017). UNSW Business School Research Paper No. 2017-10, Available at SSRN: https://ssrn.com/abstract=2947395 or http://dx.doi.org/10.2139/ssrn.2947395

Christopher Gibbs (Contact Author)

The University of Sydney - School of Economics ( email )

Rm 370 Merewether (H04)
The University of Sydney
Sydney, NSW 2006 2008
Australia

HOME PAGE: http://christopherggibbs.weebly.com

Andrey L. Vasnev

University of Sydney ( email )

Sydney, NSW 2006
Australia

HOME PAGE: http://www.econ.usyd.edu.au/staff/andreyv

Here is the Coronavirus
related research on SSRN

Paper statistics

Downloads
62
Abstract Views
616
rank
193,394
PlumX Metrics