Multivariate GARCH Models with Correlation Clustering

43 Pages Posted: 6 Feb 2010

See all articles by Mike K. P. So

Mike K. P. So

Hong Kong University of Science & Technology (HKUST) - Department of Information Systems, Business Statistics & Operations Management

Iris W.H. Yip

affiliation not provided to SSRN

Date Written: July 2009

Abstract

This paper proposes a new clustered correlation multivariate GARCH model (CC-MGARCH) that allows conditional correlations to form clusters. This model can generalize the time-varying correlation structure in Tse and Tsui (2002) by determining a natural grouping of the correlations among the series. To estimate the proposed model, we adopt Markov Chain Monte Carlo methods. Two efficient sampling schemes for drawing discrete indicators are also developed. Simulations show that these efficient sampling schemes can save substantial computation time in Monte Carlo procedures involving discrete indicators. In the applications using stock market and exchange rate data, two-cluster and three-cluster models are selected using posterior probabilities, implying that the conditional correlation equation should be governed by more than one set of decaying parameters.

Keywords: Bayesian, Markov chain Monte Carlo methods, Multivariate GARCH models, Clustering, Volatility

Suggested Citation

So, Mike K.P. and Yip, Iris W.H., Multivariate GARCH Models with Correlation Clustering (July 2009). Available at SSRN: https://ssrn.com/abstract=1548408 or http://dx.doi.org/10.2139/ssrn.1548408

Mike K.P. So (Contact Author)

Hong Kong University of Science & Technology (HKUST) - Department of Information Systems, Business Statistics & Operations Management ( email )

Clear Water Bay, Kowloon
Hong Kong

Iris W.H. Yip

affiliation not provided to SSRN ( email )

Here is the Coronavirus
related research on SSRN

Paper statistics

Downloads
168
Abstract Views
985
rank
190,493
PlumX Metrics