NATIONAL BUREAU OF ECONOMIC RESEARCH
NATIONAL BUREAU OF ECONOMIC RESEARCH

Comparing Predictive Accuracy, Twenty Years Later: A Personal Perspective on the Use and Abuse of Diebold-Mariano Tests

Francis X. Diebold

NBER Working Paper No. 18391
Issued in September 2012
NBER Program(s):   AP   EFG   IFM   TWP

The Diebold-Mariano (DM) test was intended for comparing forecasts; it has been, and remains, useful in that regard. The DM test was not intended for comparing models. Unfortunately, however, much of the large subsequent literature uses DM-type tests for comparing models, in (pseudo-) out-of-sample environments. In that case, much simpler yet more compelling full-sample model comparison procedures exist; they have been, and should continue to be, widely used. The hunch that (pseudo-) out-of-sample analysis is somehow the "only," or "best," or even a "good" way to provide insurance against in-sample over-fitting in model comparisons proves largely false. On the other hand, (pseudo-) out-of-sample analysis may be useful for learning about comparative historical predictive performance.

download in pdf format
   (369 K)

email paper

Machine-readable bibliographic record - MARC, RIS, BibTeX

Document Object Identifier (DOI): 10.3386/w18391

Published: Journal of Business & Economic Statistics Volume 33, Issue 1, 2015 Select Language​▼ Translator disclaimer Comparing Predictive Accuracy, Twenty Years Later: A Personal Perspective on the Use and Abuse of Diebold–Mariano Tests Comparing Predictive Accuracy, Twenty Years Later: A Personal Perspective on the Use and Abuse of Diebold–Mariano Tests Preview View full text Download full text View & annotate PDF Add to colwiz Library Access options DOI: 10.1080/07350015.2014.983236 Francis X. Diebolda page 1 Publishing models and article dates explained Received: 1 Sep 2014 Published online: 26 Jan 2015

Users who downloaded this paper also downloaded these:
Engel, Mark, and West w18382 Factor Model Forecasts of Exchange Rates
Diebold and Mariano t0169 Comparing Predictive Accuracy
Brunnermeier and Oehmke w18398 Bubbles, Financial Crises, and Systemic Risk
Andersen, Bollerslev, Christoffersen, and Diebold w18084 Financial Risk Measurement for Financial Risk Management
Gorton w18397 Some Reflections on the Recent Financial Crisis
 
Publications
Activities
Meetings
NBER Videos
Themes
Data
People
About

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

Contact Us