Forecast Combination Across Estimation Windows

In this article we consider combining forecasts generated from the same model but over different estimation windows. We develop theoretical results for random walks with breaks in the drift and volatility and for a linear regression model with a break in the slope parameter. Averaging forecasts over different estimation windows leads to a lower bias and root mean square forecast error (RMSFE) compared with forecasts based on a single estimation window for all but the smallest breaks. An application to weekly returns on 20 equity index futures shows that averaging forecasts over estimation windows leads to a smaller RMSFE than some competing methods.

[1]  James D. Hamilton.  Time Series Analysis , 1994 .

[2]  Todd E. Clark,et al.  Averaging Forecasts from Vars with Uncertain Instabilities , 2006 .

[3]  Everette S. Gardner,et al.  Exponential smoothing: The state of the art , 1985 .

[4]  A. Schrimpf,et al.  A reappraisal of the leading indicator properties of the yield curve under structural instability , 2010 .

[5]  Michael P. Clements,et al.  Forecasting with Breaks , 2006 .

[6]  Rejoinder to comments on forecasting economic and financial variables with global VARs , 2009 .

[7]  Tim Brailsford,et al.  Selecting the forgetting factor in subset autoregressive modelling , 2002 .

[8]  A. Timmermann.  Forecast Combinations , 2005 .

[9]  P. Perron,et al.  Computation and Analysis of Multiple Structural-Change Models , 1998 .

[10]  Stephen Gordon,et al.  Learning, Forecasting and Structural Breaks , 2008 .

[11]  Til Schuermann,et al.  Forecasting Economic and Financial Variables with Global VARs , 2007 .

[12]  Andreas Pick,et al.  Forecasting Random Walks under Drift Instability , 2008 .

[13]  M. Pesaran,et al.  Forecasting the Swiss economy using VECX models: An exercise in forecast combination across models and observation windows , 2008, National Institute Economic Review.

[14]  Davide Pettenuzzo,et al.  Forecasting Time Series Subject to Multiple Structural Breaks , 2004, SSRN Electronic Journal.

[15]  Todd E. Clark,et al.  Improving Forecast Accuracy by Combining Recursive and Rolling Forecasts , 2008 .

[16]  J. Bai,et al.  Estimation of a Change Point in Multiple Regression Models , 1997, Review of Economics and Statistics.

[17]  William A. Branch,et al.  A simple recursive forecasting model , 2006 .

[18]  R. Clemen.  Combining forecasts: A review and annotated bibliography , 1989 .

[19]  M. Hashem Pesaran,et al.  Selection of estimation window in the presence of breaks , 2007 .

[20]  J. Stock,et al.  Combination forecasts of output growth in a seven-country data set , 2004 .

[21]  F. Diebold,et al.  Comparing Predictive Accuracy , 1994, Business Cycles.

[22]  V. Solo.  The Statistical Theory of Linear Systems E. J. Hannan and Manfred Deistler John Wiley & Sons, 1988 , 1992, Econometric Theory.

[23]  E. S. Gardner.  EXPONENTIAL SMOOTHING: THE STATE OF THE ART, PART II , 2006 .

[24]  P. Perron,et al.  Estimating and testing linear models with multiple structural changes , 1995 .