Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
1155304 | Statistics & Probability Letters | 2006 | 8 Pages |
Abstract
In this note, we propose a small-sample criterion KICc for selecting vector autoregressive models. KICc is an approximately unbiased estimator of the expected Kullback's symmetric divergence. A simulation study shows that KICc provides better model order choices than the KIC criterion in small samples.
Keywords
Related Topics
Physical Sciences and Engineering
Mathematics
Statistics and Probability
Authors
Bezza Hafidi,