Template-type: ReDIF-Paper 1.0 Author-Name: Andrew B. Martinez Author-Workplace-Name: Dept of Economics and Institute for New Economic Thinking at the Oxford Martin School, University of Oxford Author-Email: Andrew.Martinez@treasury.gov Author-Name: Jennifer L. Castle Author-Workplace-Name: Magdalen College, Climate Econometrics and Institute for New Economic Thinking at the Oxford Martin School, University of Oxford Author-Email: Jennifer.Castle@magd.ox.ac.uk Author-Name: David F. Hendry Author-Workplace-Name: Nuffield College, Climate Econometrics and Institute for New Economic Thinking at the Oxford Martin School, University of Oxford Author-Email: David.Hendry@nuffield.ox.ac.uk Title: Smooth Robust Multi-Horizon Forecasts Abstract: We investigate whether smooth robust methods for forecasting can help mitigate pronounced and persistent failure across multiple forecast horizons. We demonstrate that naive predictors are interpretable as local estimators of the long-run relationship with the advantage of adapting quickly after a break, but at a cost of additional forecast error variance. Smoothing over naive estimates helps retain these advantages while reducing the costs, especially for longer forecast horizons. We derive the performance of these predictors after a location shift, and confirm the results using simulations. We apply smooth methods to forecasts of U.K. productivity and U.S. 10-year Treasury yields and show that they can dramatically reduce persistent forecast failure exhibited by forecasts from macroeconomic models and professional forecasters. Keywords: Location Shifts; Long differencing; Productivity forecasts; Robust forecasts. JEL codes: C51, C53 Length: 28 pages Creation-Date: 2021-01-14 Number: 2021-W01 File-URL: https://www.nuffield.ox.ac.uk/economics/Papers/2021/2021W01_ABMJLCDFHSR%202020-12-21.pdf File-Format: application/pdf Handle: RePEc:nuf:econwp:2101