NATIONAL BUREAU OF ECONOMIC RESEARCH
NATIONAL BUREAU OF ECONOMIC RESEARCH

Cointegration and Long-Horizon Forecasting

Peter F. Christoffersen, Francis X. Diebold

NBER Technical Working Paper No. 217
Issued in October 1997
NBER Program(s):   EFG

We consider the forecasting of cointegrated variables, and we show that at long horizons" nothing is lost by ignoring cointegration when forecasts are evaluated using standard multivariate" forecast accuracy measures. In fact, simple univariate Box-Jenkins forecasts are just as accurate. " Our results highlight a potentially important deficiency of standard forecast accuracy" measures they fail to value the maintenance of cointegrating relationships among" variables and we suggest alternatives that explicitly do so.

download in pdf format
   (857 K)

email paper

This paper is available as PDF (857 K) or via email.

Machine-readable bibliographic record - MARC, RIS, BibTeX

Document Object Identifier (DOI): 10.3386/t0217

Published: Journal of Business and Economic Statistics, Vol. 16 (1998): 450-458.

Users who downloaded this paper also downloaded these:
Diebold and Kilian w6928 Unit Root Tests Are Useful for Selecting Forecasting Models
Christoffersen and Diebold w6844 How Relevant is Volatility Forecasting for Financial Risk Management?
Cheung and Chinn w6926 Are Macroeconomic Forecasts Informative? Cointegration Evidence from the ASA-NBER Surveys
Diebold and Lopez t0192 Forecast Evaluation and Combination
Mueller and Watson w15292 Low-Frequency Robust Cointegration Testing
 
Publications
Activities
Meetings
NBER Videos
Data
People
About

Support
National Bureau of Economic Research, 1050 Massachusetts Ave., Cambridge, MA 02138; 617-868-3900; email: info@nber.org

Contact Us