This thesis aims to explore stock liquidity, a crucial attribute of financial assets, in US market. In particular, this research attempts to address a number of issues in the theoretical study of liquidity, some of which even still matters for debate. The empirical results in Chapter 3 suggest that the significance of liquidity on asset returns is time specific, in other words, the heterogeneity between liquidity components exhibits a Business Cycle effect. In particular, the liquidity risk premium is strengthened during downturns of the market conditions, as the association between the asset liquidity and return in the cross-sectional dimension is relatively stronger in the period of lower market liquidity. Besides, the analysis is carried out that focuses on the interrelationship between the market-wide liquidity components and the market dynamics, and some interesting Granger causality relationship is detected. Specifically, price impact components are Granger caused by transaction costs and trading activity, but do not Granger cause trading activity. Moreover, the Granger causality detected in this section also explains that market past performance is caused by liquidity, especially the dimensions of trading activity and price impact, and subsequently, the market-wide trading activity affects the market portfolio most recent and further performance. These findings for liquidity measures in this comparative analysis establish a significant step towards the understanding of liquidity measures in a more systematic and consistent setting, and can be a good starting point for constructing more robust liquidity measures. Based on a negative relationship between volatility of liquidity and asset returns, Chapter 4 extends this finding and provides a comparative analysis of the volatility of liquidity risk through an asset pricing framework considering several dimensions of liquidity, such as transaction cost, trading activity and price impact. The empirical findings, consistent with the literature, provide evidence of heterogeneity across various liquidity components and volatility specifications. In addition, by extracting the commonality of volatility of liquidity across individual assets via principal component analysis, the systematic components of volatility of liquidity are examined accordingly. Finally, a mimicking portfolio is constructed and used to track the systematic risk of volatility of liquidity, providing evidences that the latter is priced in asset returns. Chapter 5 studies the impact of market-wide liquidity volatility on momentum profit. It is examined by investigating whether the volatility of market liquidity dominates the market liquidity level in terms of affecting and predicting the momentum profit. Besides, it is determined that the impact is state-dependent; in particular, the impact of the fluctuation of the market liquidity on the momentum payoff is stronger when the market volatility or the illiquidity is higher. Finally, by a closer inspection of the momentum crash event in 2009, it is observed that the volatility of market liquidity increases sharply a couple of months before the crash, while stays stable during and after the crash. This thesis provides implications for investment perspective in terms of the trading strategies based on liquidity as well as momentum. For instance, the performance of the liquidity measurement is time-varying associated with market conditions. Moreover, the fluctuation of market liquidity, i.e., the volatility of liquidity, should also be considered for pricing issues. The empirical results suggest that the asset, of which the liquidity fluctuates heavily, usually has lower returns; this indication applies to six popular liquidity measures according to the empirical results. More importantly, investors could make profits by reversing the momentum trading strategy in momentum crash periods.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:679559 |
| Date | January 2015 |
| Creators | Huang, Yuping |
| Publisher | University of Glasgow |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://theses.gla.ac.uk/7036/ |
Page generated in 0.0015 seconds