Forecasting Time Series Subject to Multiple Structural Breaks
M. Hashem Pesaran
University of Southern California; Cambridge University - Faculty of Economics; CESifo (Center for Economic Studies and Ifo Institute for Economic Research); Institute for the Study of Labor (IZA)
Brandeis University - Department of Economics
Allan G. Timmermann
University of California, San Diego (UCSD) - Department of Economics; Centre for Economic Policy Research (CEPR)
CEPR Discussion Paper No. 4636
This Paper provides a novel approach to forecasting time series subject to discrete structural breaks. We propose a Bayesian estimation and prediction procedure that allows for the possibility of new breaks over the forecast horizon, taking account of the size and duration of past breaks (if any) by means of a hierarchical hidden Markov chain model. Predictions are formed by integrating over the hyper parameters from the meta distributions that characterize the stochastic break point process. In an application to US Treasury bill rates, we find that the method leads to better out-of-sample forecasts than alternative methods that ignore breaks, particularly at long horizons.
Number of Pages in PDF File: 42
Keywords: Structural breaks, forecasting, hierarchical hidden Markov Chain Model, Bayesian model averaging
JEL Classification: C11, C15, C53working papers series
Date posted: November 17, 2004
© 2013 Social Science Electronic Publishing, Inc. All Rights Reserved.
This page was processed by apollo7 in 1.735 seconds