Comparing Predictive Accuracy
Francis X. Diebold, Roberto S. Mariano · National Bureau of Economic Research · 1994
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.
Cited by 5,185 works. doi:10.3386/t0169 Open-access PDF
Authors
References
- The Advanced Theory of Statistics.
- A Simple, Positive Semi-Definite, Heteroskedasticity and Autocorrelation Consistent Covariance Matrix
- A Simple, Positive Semi-Definite, Heteroskedasticity and AutocorrelationConsistent Covariance Matrix
- Multiple Time Series
- Introduction to Mathematical Statistics.
- Spectral Analysis and Time Series
- Nonparametrics: Statistical Methods Based on Ranks
- A New Approach to the Economic Analysis of Nonstationary Time Series and the Business Cycle
- Time Series: Theory and Methods
- Forecasting Economic Time Series