Abstract
We perform an extensive Monte Carlo study of efficient method of moments (EMM) estimation of a stochastic volatility model. EMM uses the expectation under the structural model of the score from an auxiliary model as moment conditions. We examine the sensitivity to the choice of auxiliary model using ARCH, GARCH, and EGARCH models for the score as well as nonparametric extensions. EMM efficiency approaches that of maximum likelihood for larger sample sizes. Inference is sensitive to the choice of auxiliary model in small samples, but robust in larger samples. Specification tests and 't-tests' show little size distortion.
| Original language | English (US) |
|---|---|
| Pages (from-to) | 61-87 |
| Number of pages | 27 |
| Journal | Journal of Econometrics |
| Volume | 91 |
| Issue number | 1 |
| DOIs | |
| State | Published - Jul 1999 |
Funding
We are grateful to Hyung-Kwon Chung for assistance and David Tom for programming the EGARCH model in C++. The second author gratefully acknowledges financial assistance from the Ehrlich Foundation. We thank the editor, Ron Gallant, an associate editor, three anonymous referees, as well as Moshe Buchinsky, Neil Shephard, Doug Steigerwald, and seminar participants at Brown University, Yale University, University of Rochester, the NBER 1997 Summer Institute and the Econometric Society Winter Meetings, Chicago, January 1998, for comments.
Keywords
- EMM
- GMM
- Monte Carlo
- Stochastic volatility
ASJC Scopus subject areas
- Economics and Econometrics