成果1：基于实验与观测数据融合的长期因果效应估计
在很多领域研究人员通常希望了解政策的长期因果效应。例如，劳动经济学家对早期儿童教育对终身收入的影响感兴趣，在线平台对网页设计对用户长期行为的影响感兴趣。由于长期效果可能与短期效果有很大不同，准确评估长期效果对全面了解感兴趣的干预措施既困难又至关重要。
在这一项目中，我们研究通过结合短期实验数据和长期观测数据来识别和估计长期因果效应。因为实验操作挑战和观测数据中潜在混杂因素的影响，无论是单独使用短期实验数据集还是长期观测数据集都无法帮助我们正确估计长期因果效应。在这一项目中，我们利用多个短期结果的顺序结构（如下图），提出了三种新的长期因果效应识别方法。基于这些识别方法，我们开发了具有渐近保证的估计和推断方法。该项目正在《Journal of Royal Statistical Society: Series B》上审稿，目前处于大修（Major Revision）状态。
研究领域：基于观测与实验数据融合的因果推断
研究论文：Imbens, G., Kallus, N., Mao, X., & Wang, Y. (2022). Long-term causal inference under persistent confounding via data combination. arXiv preprint arXiv:2202.07234. (Major revision at Journal of Royal Statistical Society: Series B) 查看PDF
7. Lin Liu, Yuhao Wang, Root-n consistent semiparametric learning with high-dimensional nuisance functions under minimal sparsity, 2023 查看PDF
6. Xin Lu, Fan Yang, Yuhao Wang, Debiased Regression Adjustment in Completely Randomized Experiments with Moderately High-dimensional Covariates, 2023 查看PDF
5. Xin Lu, Fan Yang, Yuhao Wang, Asymptotic Theory of the Best-Choice Rerandomization using the Mahalanobis Distance, 2023 查看PDF
4. Alexandra Carpentier, Olivier Collier, Laetitia Comminges, Alexandre B. Tsybakov, Yuhao Wang, Estimation of the l2-norm and testing in sparse linear regression with unknown variance, Bernoulli, 2022 查看PDF
3. Madeline Navarro, Yuhao Wang, Antonio G. Marques, Caroline Uhler, Santiago Segarra, Joint inference of multiple graphs from matrix polynomials, Journal of Machine Learning Research, 2022 查看PDF
2. Yuhao Wang, Xinran Li, Rerandomization with Diminishing Covariate Imbalance and Diverging Number of Covariates, Annals of Statistics, 2022 查看PDF
1. Liam Solus, Yuhao Wang, Caroline Uhler, Consistency Guarantees for Greedy Permutation--Based Causal Inference Algorithms, Biometrika, 2021 查看PDF
