NATIONAL BUREAU OF ECONOMIC RESEARCH
NATIONAL BUREAU OF ECONOMIC RESEARCH

Unit Root Tests Are Useful for Selecting Forecasting Models

Francis X. Diebold, Lutz Kilian

NBER Working Paper No. 6928
Issued in February 1999
NBER Program(s):   EFG   AP

We study the usefulness of root tests as diagnostic tools for selecting forecasting models. Difference stationary and trend stationary models of economic and financial time series often imply very different predictions, so deciding which model to use is tremendously important for applied forecasters. Forecasters face three choices: always difference the data, never difference, or use a unit-root pretest. We characterize the predictive loss of these strategies for the canonical AR(1) process with trend, focusing on the effects of sample size, forecast horizon, and degree of persistence. We show that pretesting routinely improves forecast accuracy relative to forecasts from models in differences, and we give conditions under which pretesting is likely to improve forecast accuracy relative to forecasts from models in levels.

download in pdf format
   (906 K)

email paper

This paper is available as PDF (906 K) or via email.

Machine-readable bibliographic record - MARC, RIS, BibTeX

Document Object Identifier (DOI): 10.3386/w6928

Published: Diebold, Francis X. and Lutz Kilian. "Unit-Root Tests Are Useful For Selecting Forecasting Models," Journal of Business and Economic Statistics, 2000, v18(3,Jul), 265-273. citation courtesy of

Users who downloaded this paper also downloaded these:
Christoffersen and Diebold t0217 Cointegration and Long-Horizon Forecasting
Schwert t0073 Tests For Unit Roots: A Monte Carlo Investigation
Elliott, Rothenberg, and Stock t0130 Efficient Tests for an Autoregressive Unit Root
Christiano and Eichenbaum w3130 Unit Roots in Real GNP: Do We Know, and Do We Care?
Campbell and Perron t0100 Pitfalls and Opportunities: What Macroeconomists Should Know About Unit Roots
 
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