| Article ID | Journal | Published Year | Pages | File Type |
|---|---|---|---|---|
| 1153004 | Statistics & Probability Letters | 2010 | 7 Pages |
Abstract
Event forecast is a possibly autocorrelated binary time series. A new test for its timing ability is based on the correlation between the discrete autoregressions of the event forecast and the event time series. The new test outperforms the existing market timing tests.
Related Topics
Physical Sciences and Engineering
Mathematics
Statistics and Probability
Authors
Cheng Chou, Chia-Shang J. Chu,