Haochen Li is currently a Ph.D. candidate in the Department of Informatics at King's College London. He holds a bachelor's degree in Electronics and Electrical Engineering from the University of Edinburgh. Soon after his bachelor’s studies he enrolled and completed his master studies in the Department of Mathematics at the London School of Economics. Prior to starting his Ph.D. research, Haochen was a institutional proprietary trader, trading futures derivatives of commodities, fixed income and short-term interest rates. Haochen's research interests include econophysics, market microstructure, and AI for finance.
High-frequency order book data has drawn increasing attention both in academia and industry with the growing computing power. Most current literatures on electronic trading merely focus on the price dynamics and neglect the order book data. However, the activities of orders from the data are the most granular microstructure information for the financial market. The orders constitute the order book and further the market dynamics hence it is natural to study their mass activities.
As current market measures including price, volatility, bid-ask spread, and volume are merely plain description on the results of market dynamics, the question is - how to find the intrinsic reasons that cause the changes on these market measures?
The market dynamics are driven by the order dynamics on the microscopic scale, hence this research will propose a method modelling the specific process and statistical laws that the order dynamics evolve the market dynamics. Statistical physics and computational fluid dynamics techniques will be employed to model high-frequency trading data. It is aimed to discover the related probability distribution function for describing the changes in the states of the limit order book and further the market and price dynamics. Bayesian neural network and reinforcement learning will be also introduced based on the labelled data.
This research will discover the hierachy of complex systems for financial markets, which is an epitime of our social economy. It will provide an example on applying statistical physics and machine learning on the complex systems with social data science.