Font Size: a A A

Volatility Modeling With Markov Regime Switching And Its Applications

Posted on:2004-06-27Degree:MasterType:Thesis
Country:ChinaCandidate:M Y GuoFull Text:PDF
GTID:2156360122487546Subject:Technical Economics and Management
Abstract/Summary:PDF Full Text Request
The thesis is mainly focused on volatility modeling with markov regime switching and its applications in Chinese stock markets. The main contexts of this thesis are as follows:The main types of volatility models with structure changes, including subsection modeling, GARCH models and SV models with markov regime switching and heavy-tail regime switching model are introduced. Both the definition and the formula of stochastic fluctuations and abnormal fluctuations are given. Then we analyse the four main factors influencing the fluctuations of stock prices. In addition, we apply the markov regime switching algorithm to detect the abnormal fluctuations of stock prices in Shanghai stock market.We put forward the DDMRS-GARCH model (duration dependence markov regime switching GARCH model). In the DDMRS-GARCH model, the transition probability of the volatility state variable is not only related to the volatility state, but also to the duration of the volatility state. We compare the efficiencies of the DDMRS-GARCH model and the MRS-GARCH model from three aspects using the daily returns data in Shanghai stock markets. We find that the DDMRS-GARCH model is superior to the MRS-GARCH model. In Chapter Six, we combine the MRS-GARCH model with CCAPM. In this model, the beta is changing with the volatility state. The empirical tests show that CCAPM under the MRS-GARCH model with changing beta is superior to CCAPM under the GARCH model with constant beta.This thesis is part of the research on the volatility persistence of multivariate time series and its applications in financial analysis which is funded by National Science Fundation(NO.70171001).
Keywords/Search Tags:abnormal fluctuations, DDMRS-GARCH model, MRS-GARCH model, CCAPM, changing beta
PDF Full Text Request
Related items