| Summary: | This thesis investigates the impact of technological and regulatory changes on UK equity market microstructure, and the implications of these changes for policy makers, regulators and market participants. In the first analysis, we model the execution performance of two popular volume participation algorithms. We compare the in-sample fit and out-of-sample predictive ability of two alternative models of execution costs, and find that the non-linear model provides a better fit than the linear model. We also examine the relative importance of different order-specific, stock-specific and market-specific variables in explaining the execution performance of these algorithms. We show that execution risk for volume participation algorithms comprises not just price risk, but also risk due to uncertain trading volumes. The growth in high·frequency trading has been one of the most significant developments in the equity trading landscape, and following a number of market mishaps; has also caught the attention of regulators. In the second analysis, we examine the intraday behavior of high-frequency traders and their impact on market quality. We first observe that high-frequency trading strategies differ significantly from each other in terms of the level of liquidity provision. We next explore the impact of different high-frequency trading strategies on price discovery and temporary- deviations from equilibrium values (noise). We find that all high-frequency traders have a larger contribution towards price discovery m iv ABSTRACT and noise than other traders in the market, thereby amplifying both the beneficial and detrimental components of price volatility. Finally, in the last analysis, we revisit issues related to the liquidity characteristics of limit order markets after Market in Financial Instruments Directive was operationalised in the European Union. We find that the top of the London Stock Exchange's limit order book is extremely thin, and the slope of the limit order book is steep near the top. We further observe that the limit order book contains significant information about future short-term price changes, especially for the less liquid stocks, and this information has economic value in an algorithmic trading environment.
|
|---|
