Research
Research Experience
Robust Sensitivity Analysis for Quantiles of Individual Treatment Effects in Paired Observational Studies
June 2021 - Present, advised by Prof. Xinran Li, University of Chicago
Generalized the conventional sensitivity analysis to an robust sensitivity analysis for quantiles of individual effects with quantiles of the hidden biases in matched study.
Implemented the algorithm in R and developing R packages.
Sensitivity Analysis for Quantiles of Hidden Biases in Matched Observational Studies
June 2020 - June 2021, advised by Prof. Xinran Li, University of Chicago
Generalized the conventional sensitivity analysis to deal with general quantiles of the hidden biases from all matched sets, which are more robust than the maximum. The framework works for general outcomes, general matched studies and general test statistics.
Demonstrate through simulations and implications that the proposed sensitivity analysis for all quantiles of hidden biases is simultaneously valid and is thus a free lunch added to the conventional sensitivity analysis.
Provided an R package for the proposed method.
Recommendation System on Traditional Chinese Medicine Big Data
June 2017 - June 2018, advised by Prof. Xiaohua Zhou, Peking University
Implemented recommendation systems in Python through neighborhood-based and model-based collaborative filtering, and deep learning based techniques such as item embedding, feedforward networks, auto-encoders.
Constructed hybrid recommendation system based on DeepFM and neighborhood-based model and tested on real data.
Publications
Sensitivity Analysis for Quantiles of Hidden Biases in Matched Observational Studies.
Dongxiao Wu, Xinran Li. Under reject with resubmission at Journal of the American Statistical Association. [arXiv] [R package]
Robust Sensitivity Analysis for Quantiles of Individual Treatment Effects in Paired Observational Studies.
Dongxiao Wu, Xinran Li. To be submitted.