| Financial markets data often exhibits volatility clustering. The sources of volatility clustering may be the nominal interest rate, dividend yield, oil price, business cycle, money supply and information patterns, etc. Therefore, if information comes in clusters, the returns would exhibit clustering accordingly. Moreover, the participants need time to digest the information shocks due to their prior beliefs and then resolve their expectation difference, market also would result in volatility clustering (Takatoshi, Engle, & Lin, 1990). Time varying volatility is more common than constant volatility. Therefore, the accurate modeling of time varying volatility is important in financial markets for the purpose of asset pricing, and understanding the time series dynamics of financial data. Generally, ARMA models can be used to model the conditional variance based on the past process, but a more desirable approach is to use GARCH family models, which are widely used in finance world and researchers often use it to analyze time varying volatility. The main purpose of this research is to investigate the volatility of Chinese equity markets (Mainland Shanghai and Shenzhen Markets and Special administrative region Hong Kong Market) with GARCH type models. In this study, we are trying to capture the volatility dynamics in these three markets and then test the relationship among them. After these tests, we explore the volatility transmission within one country and between the three markets. |
