You are not currently logged in.
Access JSTOR through your library or other institution:
Comparing Predictive Accuracy
Francis X. Diebold and Roberto S. Mariano
Journal of Business & Economic Statistics
Vol. 13, No. 3 (Jul., 1995), pp. 253-263
Stable URL: http://www.jstor.org/stable/1392185
Page Count: 11
Preview not available
We propose and evaluate explicit tests of the null hypothesis of no difference in the accuracy of two competing forecasts. In contrast to previously developed tests, a wide variety of accuracy measures can be used (in particular, the loss function need not be quadratic and need not even be symmetric), and forecast errors can be non-Gaussian, nonzero mean, serially correlated, and contemporaneously correlated. Asymptotic and exact finite-sample tests are proposed, evaluated, and illustrated.
Journal of Business & Economic Statistics © 1995 American Statistical Association