Xiaojie Mao

Assistant Professor

Tsinghua University

Biography

I am an assistant professor in Management Science and Engineering at Tsinghua University. I obtained my PhD degree in Statistics from Cornell Department of Statistics and Data Science, advised by Nathan Kallus and Madeleine Udell. Prior to my graduate school, I studied Mathematical Economics at Wuhan University.

Broadly, I work on data-driven decision making and causal inference, with a particular focus on developing statistical methods for problems arising in these areas. I am interested in developing reliable methodology for decision making and statistical inference when faced with practical data challenges such as measurement errors, omitted variables, missing data, among others.

Here is my Tsinghua webpage in Chinese.

Interests

Data-driven decision-making
Causal Inference
Statistical Learning

Education

PhD in Statistics, 2016 - 2021

Cornell University
BS in Mathematical Economics, 2012 - 2016

Wuhan University

Teaching

I am a teaching assistant for the following courses at Cornell University:

ORIE 5750/CS 5785 Applied Machine Learning, 2019 FALL.
BTRY 6010/ILRST 6100 Statistical Methods I, 2016 FALL, 2017 FALL.
STSCI 5110/ILRST 5110 Statistical Methods for Social Science, 2018 Spring.

Xiaojie Mao

Assistant Professor

Tsinghua University

Biography

Interests

Education

Preprint

Publications

Talks

Stochastic Optimization Forests

Smooth contextual bandits: bridging the parametric and non-differentiable regret regimes

Localized Debiased Machine Learning: Efficient Inference on Quantile Treatment Effects and Beyond

Smooth contextual bandits: bridging the parametric and non-differentiable regret regimes

Smooth contextual bandits: bridging the parametric and non-differentiable regret regimes

Partial identification for unmeasured confounding with applications in causal inference and algorithmic fairness

Partial identification for unmeasured confounding with applications in causal inference and algorithmic fairness

Smooth contextual bandits: bridging the parametric and non-differentiable regret regimes

Partial identification for unmeasured confounding with applications in causal inference and algorithmic fairness

Causal Inference with Noisy and Missing Covariates via Matrix Factorization

Assessing Algorithmic Fairness with Unobserved Protected Class Using Data Combination

Interval Estimation of Individual-Level Causal Effects Under Unobserved Confounding

Assessing Disparity When Protected Class Is Unobserved

Assessing Outcome Disparity with Proxy for Unobserved Protected Class

Teaching

Contact