Small-sample properties of dimensionality statistics for fitting VAR models to aggregate economic data: A Monte Carlo study |
| |
Authors: | Gerald Nickelsburg |
| |
Affiliation: | University of Southern California, Los Angeles, CA 90089-0035, USA |
| |
Abstract: | The analysis of aggregate economic phenomena by VAR's as suggested by Sims often results in a small sample relative to the number of estimated parameters. Since the model is identified by a dimensionality criterion, the small-sample properties of available criteria are important. This paper presents a study of small-sample properties for six criteria with Monte Carlo methods. It is found that no criterion performs well, and that underfitting of models may be quite common. |
| |
Keywords: | |
本文献已被 ScienceDirect 等数据库收录! |
|