This thesis is an empirical study of the volatility and correlation in financial markets, and consists of two parts: The first part is on econometric modeling of the volatility and correlation (Chapter 1). The second part is on the pricing implication of the correlation and volatility as risk factors (Chapter 2 and 3). The thesis begins with proposing a regime-switching multivariate GARCH model of volatilities and correlation. We incorporate the Markov-switching mechanism into the Constant Conditional Correlation model (CCC). The proposed model allows us to capture the different dynamics in both the volatilities and correlations in different regimes. It is particularly useful in examining the contemporaneous relationship between the unobservable volatility and correlation processes. We apply our model to the stock market index paired with two bond market indexes. Then in the second chapter, we estimate the risk premium for the average correlation in the cross-section of the US stock market. The average correlation is the cross-sectional average of the correlations between each pair of stocks in the stock market. We find there is a negative and statistically significant risk premium for the average correlation controlling for other factors such as Fama-French's SMB and HML as well as the liquidity factor and momentum factor. The third chapter focuses on the risk-return relation using a set of variance-related risk measures. We combine two lines of literature both of which find significant forecasting power of some risk measures for stock market returns: variance risk premium literature and studies on average variance-correlation decomposition. We illustrate how these two approaches can be related to each other, and empirically evaluate the relative importance of two approaches.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:645829 |
| Date | January 2009 |
| Creators | Chen, Runquan |
| Publisher | London School of Economics and Political Science (University of London) |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://etheses.lse.ac.uk/2354/ |
Page generated in 0.0018 seconds