Forecasting Time Series Subject to Multiple Structural Breaks

42 Pages Posted: 17 Nov 2004

See all articles by M. Hashem Pesaran

M. Hashem Pesaran

University of Southern California - Department of Economics

Davide Pettenuzzo

Brandeis University - International Business School

Allan Timmermann

UCSD ; Centre for Economic Policy Research (CEPR)

Multiple version iconThere are 2 versions of this paper

Date Written: September 2004

Abstract

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.

Keywords: Structural breaks, forecasting, hierarchical hidden Markov Chain Model, Bayesian model averaging

JEL Classification: C11, C15, C53

Suggested Citation

Pesaran, M. Hashem and Pettenuzzo, Davide and Timmermann, Allan, Forecasting Time Series Subject to Multiple Structural Breaks (September 2004). Available at SSRN: https://ssrn.com/abstract=621549

M. Hashem Pesaran

University of Southern California - Department of Economics ( email )

3620 South Vermont Ave. Kaprielian (KAP) Hall 300
Los Angeles, CA 90089
United States

Davide Pettenuzzo

Brandeis University - International Business School ( email )

Mailstop 32
Waltham, MA 02454-9110
United States

Allan Timmermann (Contact Author)

UCSD ( email )

9500 Gilman Drive
La Jolla, CA 92093-0553
United States
858-534-0894 (Phone)

HOME PAGE: http://rady.ucsd.edu/people/faculty/timmermann/

Centre for Economic Policy Research (CEPR)

London
United Kingdom