Quantcast
Channel: The Stata Blog - Latest Comments
Viewing all articles
Browse latest Browse all 1128

Re: Tests of forecast accuracy and forecast encompassing

$
0
0

The Clark and McCracken statistic tests competing nested parametric models. Diebold and Mariano's test is more general, in that the models need not be related. You could compare, for instance, an AR(1) forecast to a regression model's forecast (or a nonlinear model's forecast) of a given series. In Stata, ssc describe dmariano.


Viewing all articles
Browse latest Browse all 1128

Trending Articles