Improving Forecast Accuracy by Combining Recursive and Rolling Forecasts

By Todd E. Clark, Michael W. McCracken. 2008

204 citations recorded by the source corpus.

This paper presents analytical, Monte Carlo, and empirical evidence on combining recursive and rolling forecasts when linear predictive models are subject to structural change. Using a characterization of the bias-variance tradeoff faced when choosing between either the recursive and rolling schemes or a scalar convex combination of the two, we derive optimal observation windows and combining weights designed to minimize mean square forecast error. Monte Carlo experiments and several empirical examples indicate that combination can often provide improvements in forecast accuracy relative to forecasts made using the recursive scheme or the rolling scheme with a fixed window width.

Subfields: Economics and Econometrics, Management Science and Operations Research

View Paper: https://www.semanticscholar.org/paper/4cc50c764df693e963a6c709ff233bca0654584a

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Economics and Econometrics

In Management Science and Operations Research

Citations in the corpus

Advances in Forecasting Under Instability (2012, unknown): “Clark and McCracken (2009) show that the optimal window to use in the rolling scheme is R = t b (so that the optimal window uses only data after the break) and that the forecast that minimizes the MSFE is a weighted average of the rolling and recursive parameter estimates: t b rec;t + (1 t ) b…”
Forecast combination and the Bank of England’s suite of statistical forecasting models (2008, unknown): “In an interesting paper, Clark and McCracken (2004) report that the accuracy of forecasts from a given model can be improved by combining forecasts from a model estimated with the full sample of available data with forecasts from a model estimated with a rolling sample of data. Clark and McCracken (”
Forecasting Random Walks Under Drift Instability (2008, background): “More closely related to our approach is the suggestion by Clark and McCracken (2004) that averaging expanding and rolling windows can be useful for forecasting when faced with structural breaks.”
Nowcasting GDP in Real Time: A Density Combination Approach (2014, unknown): “Lately, increased interest has also been given to mixed data sampling (MIDAS) models (see, among others, Clements and Galvão (2008, 2009), Ghysels and Wright (2009) and Kuzin et al.”
Out-of-Sample Forecast Tests Robust to the Choice of Window Size (2011, unknown): “This article is closely related to the works by Pesaran and Timmermann (2007) and Clark and McCracken (2009), and more distantly related to Pesaran, Pettenuzzo, and Timmermann (2006) and Giacomini and Rossi (2010).”
Forecast Combination Across Estimation Windows (2011, unknown): “There is a sizeable literature on the merits of combining forecasts obtained from different models, reviewed by Clemen (1989), Stock and Watson (2004), and, more recently, by Timmermann (2006).”