Background
I earned a BSc in Mathematics and Applied Mathematics from Huazhong University of Science and Technology in 2020. I then obtained MSc degrees in Financial Math and Statistics from King’s College London and the University of Warwick. During my Financial Math degree, I specialised in machine learning for solving PDEs and focused on Deep Pricing of American options in the CEV model. In my Statistics degree, I developed a strong interest in Bayesian Statistics and MCMC methods, which deepened my desire for further research. I joined STOR-i in 2023, and after completing one year of the MRes program, I have been conducting PhD research on Exchangeable Particle Gibbs methods for reaction networks under the supervision of Prof. Chris Sherlock and Dr. Lloyd Chapman.
I appreciate the focus on teamwork and problem-solving within STOR-i. Moreover, I eagerly anticipate the opportunity to collaborate with industry, allowing me to witness how theoretical concepts and methodologies are applied in practical contexts. Additionally, the prospect of delving into diverse topics during the MRes year, enabling me to pinpoint my specific research interests, excites me. Furthermore, my year of PhD experience has given me a clearer understanding of what good research involves and has strengthened my enthusiasm for pursuing it. I now eagerly anticipate the moment when I can embark on my own independent research journey.
Experience
My passion for Statistics and Financial Mathematics initiated from my studying experience in Financial Mathematics at King’s College London. The knowledge that I gained during the Risk Neutral Valuation module made me realize that one of the most important and basic models in pricing of options, the Black-Scholes-Merton model, is based on the assumptions of ideal market conditions. However, the real financial market may not satisfy these assumptions, so I am interested in option pricing outside of the Black-Scholes-Merton model. Besides, my experience at the Econophysics module inspired my interest in applying statistical methods in
finance. Precisely, by looking at historical data, it can be found that the probability distribution of equity prices and their tails show that they are not Gaussian but power law, which contradicts the assumption of a perfect market in many financial models.
During the final project at King’s College London, I built a deep neural network to approximate the pricing function mapping from the parameters like stock prices, strike prices and time to maturity to the prices of American put options computed by the Finite Difference Method and did model calibration using this neural network. Then I learned that machine learning can be used to enable the rapid computations of option prices, which is further essential to model calibration and the computation of risk of option portfolios. In conclusion, my current research interests mainly involve the application of machine learning and other statistical methods in finance.
To further explore these interests, I began my second MSc program in Statistics at the University of Warwick. The experience here not only strengthened my basic knowledge of statistics like Linear Regression, Probability and Hypothesis Testing, but also widened my insight into simulation-based techniques like the Monte Carlo method. The content of Monte Carlo module sparked my new interest in researching Monte Carlo Markov Chains (MCMC) and a variety of sampling algorithm like the Gibbs Sampler. Therefore, besides my interest in Financial Mathematics, I would like to leave some space for more possible research interests in MCMC.
My experience in STOR-i has been among the most enjoyable and meaningful times of my studies. In particular, I have gained a deep sense of achievement and satisfaction from conducting research on MCMC and SMC topics with my supervisors. Through this experience, I have discovered where my true passion lies. I have not only learned how to analyse problems effectively, but also developed a clearer sense of which research questions are worth pursuing. This shift in perspective has shaped the way I approach my academic work and deepened my engagement with research. I have gained a thorough understanding of particle filters, particle MCMC methods, and techniques such as ancestor sampling to address particle degeneracy. In addition, I have improved my coding skills, particularly in parallel computing, integrating R with C++, and writing more efficient code.