Liu, Bo2024-06-252024-06-252024https://hdl.handle.net/1828/16669The landscape of financial markets has undergone a profound transformation with the advent of high-frequency trading (HFT), fundamentally altering traditional notions of market dynamics. The dissertation explores the profound transformation of financial markets by HFT, examining its impact on trading speed, systematic risk, and market sentiment. Traditional methodologies are questioned, leading to an exploration of market microstructure and the integration of technologies like machine learning. Chapter 1 sets the stage by discussing the evolving nature of financial markets and the necessity of adapting research methodologies. Chapter 2 analyzes the regulation of trading speed, revealing trade-offs in market liquidity and price discovery. Chapter 3 focuses on detecting and mitigating mini flash crashes, leveraging machine learning to develop an Early Warning System. Chapter 4 examines the efficacy of speed bump mechanisms in reducing mini flash crashes, highlighting both benefits and unintended consequences. Overall, the dissertation enhances understanding of HFT’s effects on market dynamics, risk management, and regulation. This dissertation contributes to a deeper understanding of the ramifications of high-frequency trading on market dynamics, risk management strategies, and regulatory paradigms.enAvailable to the World Wide Webspeed bumpprice discoverymarket liquiditymarket efficiencymachine learningmini flash crashThesis