Welcome to my website! I am a fifth year Ph.D student in Decision Sciences at Fuqua School of Business, Duke University. I am very fortunate to be advised by Prof. Alexandre Belloni currently. Previously, I obtained my master in Statistics at Boston University. During the time at BU, it is my pleasure to be advised by Prof. Daniel Sussman and Prof. Konstantinos Spiliopoulos to explore graph matching in network and system risk. Prior to my master, I received my Bachlor degrees in Economics and Software Engineering in Guangdong University of Foreign Studies. Before going to college, I spent three amazing years at Guiyang No.1 high school.

Research Interests

My research interest lies in empirical process, methodology and applications of causal inference. In particular, I am interested in the convergence rate of error measures for machine learning algortihms by using tools in empirical process.

I am currently working on developing methodology for estimating treatment effect by matching method under network interference with Prof. Alexandre Belloni and Prof. Alexander Volfovsky. Meanwhile, I am working on estimating causal effect of name order on election ballot with clustered data with Prof. Alessandro Arlotto.

What’s new

• 10/24/2021: Presentation “Estimating Direct Causal Effect under Hierarchical Interference on Networks in Observational Studies” in session “Cutting-Edge Methods for Data-Driven Decision-Making” in INFORMS.

• 08/10/2021: Presentation “On the Causal Effect of Flipping the Party Order of Candidates in North Carolina’s Nonpartisan Elections” in session “Social Statistics Speed Session” in JSM.

Publications

1. Fei Fang, Daniel L. Sussman, Vince Lyzinski. “Tractable Graph Matching via Soft Seeding” arXiv preprint:1807.09299 (2018). [arXiv]
2. Fei Fang, Yiwei Sun, Konstantinos Spiliopoulos. “The Effect on Heterogeneity on Flocking Behavior and System Risk” Statistics & RiskModeling, Volume 34: Issue 3-4 (2017). [Link]

Working Paper

1. A Uniform Regret Bound for the Quantile Regression Tree Process in Online Learning, with Alexandre Belloni.
2. Estimating Direct Causal Effect under Hierarchical Interference on Networks in Observational Studies, with Alexandre Belloni and Alexander Volfovsky.
3. On the Causal Effect of Flipping the Party Order of Candidates in North Carolina’s Nonpartisan Elections, with Alessandro Arlotto, Alexandre Belloni and Saša Pekeč.