Cointegration and Long-Horizon Forecasting
Electronic Access:
Free Download. Use the free Adobe Acrobat Reader to view this PDF file
Summary:
Imposing cointegration on a forecasting system, if cointegration is present, is believed to improve long-horizon forecasts. Contrary to this belief, at long horizons nothing is lost by ignoring cointegration when the 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.
Series:
Working Paper No. 1997/061
Subject:
English
Publication Date:
May 1, 1997
ISBN/ISSN:
9781451848137/1018-5941
Stock No:
WPIEA0611997
Pages:
30
Please address any questions about this title to publications@imf.org