Prior to joining PKU, I obtained my Ph.D. in Applied Mathematics from MIT in 2015, under the supervision of Andrew W. Lo. I received my bachelor's degrees in Mathematics and Applied Mathematics, and Economics (double degree) from Peking University in 2011. I also worked at Google and Goldman Sachs in the past.
欢迎申请北京大学2023年博雅博士后项目. I am looking for PostDocs on a rolling basis. Please drop me an email with your CV if you are interested in working together with me.
2023: I am organizing three exciting sessions in upcoming conferences. Please join us if you are around!
“Information and Market Microstructure” at the 2023 INFORMS Annual Meetings
2023: Our working paper “On Consistency of Signatures Using Lasso” is now on arxiv.
We revisits the consistency issue of Lasso regression for the signature transform, both theoretically and numerically.
2022: Our working paper “Optimal Impact Portfolios with General Dependence and Marginals” is now on SSRN.
We develop the impact portfolio construction framework first proposed in Lo and Zhang (2021) to allow for general dependence between impact and returns, as well as general marginals of the return distributions.
2023: Congratulations to Chaoyi Zhao (PhD student) who has won Second place in the Best Paper Prize for Young Scholars at the Annual Conference of the Operations Research Society of China (Financial Engineering and Risk Management Branch).
2023: Our working paper “Quantifying the Returns of ESG Investing: An Empirical Analysis with Six ESG Metrics” is now on SSRN.
We quantify the financial performance of ESG portfolios in the U.S., Europe, and Japan, based on data from six major ESG rating agencies. We propose several statistical and voting-based methods to aggregate individual ESG ratings, the latter based on the theory of social choice. Overall, we find that there exists a significant signal in ESG rating scores that can be used for portfolio construction despite their noisy nature.
2023: Our working paper “Estimating Market Liquidity from Daily Data: Marrying Microstructure Models and Machine Learning” is now on SSRN.
We apply (interpretable) machine learning to estimate market liquidity by combining human-engineered liquidity proxies based on microstructure models and widely available low-frequency data. Combining human-engineered proxies and the raw low-frequency data achieves cross-sectional correlations of over 0.95 and time-series correlations of over 0.70 between model estimates and the ground-truth average spread.
How do you measure the financial reward (or cost) of investing towards carbon neutrality?
We study the performance of green portfolios in both the US and Chinese markets, constructed using a broad range of climate-related environmental metrics.
2022: Our working paper “Spectral Volume Models: High-Frequency Periodicities in Intraday Trading Activities” is now on SSRN.
We develop a spectral model for high-frequency intraday trading volumes, and find very strong and consistent periodicities at a few round-second / round-minute frequencies, across a large panel of stocks in both US and China. We study why they happen and how they are useful for volume prediction.
2022: Our working paper “High-Frequency Liquidity in the Chinese Stock Market: Measurements, Patterns, and Determinants” is now on SSRN.
We study a range of high-frequency liquidity measures in the Chinese stock market using limit order book data.
2022: Our working paper “Channel and Spatial Attention CNN: Predicting Price Trends from Images” is now on SSRN.
We propose an attention-based convolutional neural network for price trend prediction that takes arbitrary images constructed from financial time series data as input.
The model achieves good out-of-sample performance and learns meaningful technical patterns that are interpretable by humans.
How do bias, polarization, and other challenges to collective intelligence happen? We propose ways to prevent such failures by nudging the “madness of mobs” towards the “wisdom of crowds” through shifts in the environment.
2022: Our working paper “Social Contagion and the Evolutionary Survival of Diverse Investment Styles” is now on SSRN.
We model the contagion of investment ideas in a multi-period setting, and show that a greater diversity in investment styles are able to survive compared to what traditional theory predicts.
2021: Our working paper “Quantifying the Impact of Impact Investing” is now on SSRN.
We propose a quantitative framework for assessing the financial impact of any form of impact investing, including SRI, ESG, and even the Gamestop Phenomenon.
2021: Our working paper “Explainable Machine Learning Models of Consumer Credit Risk” is now on SSRN.
We create machine learning (ML) models to forecast home equity credit risk for individuals using a real-world dataset and demonstrate methods to explain the output of these ML models to make them more accessible to the end-user.
Are we “rational” in financial decision making? We conduct an experiment with real monetary payoffs to show that people engage in probability matching, also known as the “matching law” or Herrnstein’s Law.
I co-organize the regular Seminar series in Financial Mathematics at Peking University.
I am looking for students and PostDocs to work with me on a variety of exciting projects. If you have a strong background in math / statistics / machine learning / FinTech / quantitative finance, please drop me an email with your CV.