On the multivariate EGARCH model

Abstract

In this aticle, the extension of Nelson's (1991) univariate EGARCH model to the multivariate version has been reexamined and compared with the existing one given by Koutmos and Booth (1995). The magnitude and sign of standardized innovations have been constrained in Koutmos and Booth's multivariate EGARCH model, but not in the actual multivariate EGARCH model. The constraints imposed on Koutmos and Booth's EGARCH model may lead to inaccurate parameter estimates. Since the actual multivariate EGARCH model obtained is more general, and can produce more accurate inferential results, we suggest that the actual multivariate EGARCH model be used in future financial empirical studies.

Acknowledgements

We thank three anonymous referees and the editor for helpful comments and suggestions. This research was partially supported by grant NSC 94-2416-H-009-012 from the National Science Council of Taiwan, R.O.C.