Modelling extreme returns in Chinese stock market using extreme value theory and copula approach

Hussain, S 2016, Modelling extreme returns in Chinese stock market using extreme value theory and copula approach, Doctor of Philosophy (PhD), Graduate School of Business and Law, RMIT University.

Document type: Thesis
Collection: Theses

Attached Files
Name Description MIMEType Size
Hussain.pdf Thesis application/pdf 2.16MB
Title Modelling extreme returns in Chinese stock market using extreme value theory and copula approach
Author(s) Hussain, S
Year 2016
Abstract The Chinese stock market has unique features that make it a challenging and interesting research topic. It is one of the biggest stock markets in the world in terms of capitalization yet it is still considered to be an emerging market. This market is very volatile and thus displays some extreme behaviour. Extreme movements in share returns rarely occur. However, they can have devastating consequences. Thus, it is important for investors, speculators and risk managers to comprehend extreme movement events in stock markets. To this end, we employ Extreme Value Theory (EVT) and copulas in this study.

First, we are concerned with the distribution of the extreme daily returns of the Chinese stock market. Generalized Extreme Value (GEV), Generalized Logistic (GL) and Generalized Pareto (GP) distributions are three well-known distributions in extreme value theory. These distributions are used to identify the distribution which is best fitted with the extreme returns. Our results indicate that the GL distribution is a better fit for the minima series and the GEV distribution is a better fit for the maxima series based on daily returns in the Chinese stock market from 1991 to 2013. This is in contrast to the previous studies, such as the one in the US and Singapore stock markets. This finding also considers extreme events that occurred and could potentially impact on the Chinese stock market such as the introduction of stock movement restriction and the Global Financial Crisis (GFC). Our results are robust regardless of these extreme events.

Second, this study explores the dependence structure between the Chinese stock market and other major stock markets. This study reveals that the Chinese stock market seems to be more strongly integrated with stock markets in Australasia than Europe or the United States. It is shown that the dependence between Chinese stock market and those of other stock markets is stronger during the crisis period than a normal period. This study also suggests that not much benefit can be gained during a downturn in portfolio diversification across the pairs of stock markets considered.

Third, this study examines the dependence structure between the stock markets in the Greater China Economic Area (GCEA) including mainland China, Hong Kong and Taiwan. These stock markets have become more and more important in recent years. However, little is known about the dependence structure between these markets. This research reveals that the dependence between all pairs of GCEA stock markets is strong. As expected, the Shanghai-Shenzhen pair has the strongest dependence (overall, lower tail and upper tail) among all pairs considered. This is followed by the Hong Kong-Taiwan pair. This study finds that diversification is effective for two pairs (Shenzhen-Hong Kong and Hong Kong-Taiwan) in the context of negative market extreme events.

Overall, this study reveals some important results regarding the extreme behaviour in the Chinese stock market. This study also demonstrates that combining EVT and copula can provide an effective way to understand extreme behaviour in stock markets. This approach can lead to incremental insights to the conclusions based on the normality assumption. The outcome of this study can have important implications for policy-making and risk management.
Degree Doctor of Philosophy (PhD)
Institution RMIT University
School, Department or Centre Graduate School of Business and Law
Subjects Finance
Investment and Risk Management
Applied Statistics
Keyword(s) Chinese stock market
Extreme returns
Risk management
Extreme Value Theory
Dependence structure
Global financial crisis (GFC)
Version Filter Type
Access Statistics: 256 Abstract Views, 475 File Downloads  -  Detailed Statistics
Created: Fri, 04 Nov 2016, 12:17:07 EST by Keely Chapman
© 2014 RMIT Research Repository • Powered by Fez SoftwareContact us