Date Added: Dec 2010
The authors examine how to forecast after a recent break. They consider monitoring for change and then combining forecasts from models that do and do not use data before the change; and robust methods, namely rolling regressions, forecast averaging over different windows and Exponentially Weighted Moving Average (EWMA) forecasting. They derive analytical results for the performance of the robust methods relative to a full-sample recursive benchmark. No clear ranking emerges under deterministic breaks. In Monte Carlo experiments forecast averaging improves performance in many cases with little penalty where there are small or infrequent changes. Similar results emerge when they examine a large number of UK and US macroeconomic series.