报告时间：2025年6月19号 15：00-16：00

报告地点： 国产自拍 106

报告题目: A New and Efficient Debiased Estimation of General Treatment Models by Balanced Neural Networks Weighting

报告摘要: Estimation and inference of treatment effects under unconfounded treatment assignment often suffer from bias and the `curse of dimensionality' due to the nonparametric estimation of nuisance parameters for high-dimensional confounders. Although debiased state-of-the-art methods have been proposed for binary treatments under particular treatment models, directly extending them to general treatment models can lead to computational complexity and unstable estimation. We propose a balanced neural networks weighting (BNNW) method for general treatment models, which leverages deep neural networks (DNN) to alleviate the curse of dimensionality while retaining optimal covariate balance through calibration, thereby achieving debiased and robust estimation. Our method accommodates a wide range of treatment models, including average, quantile, distributional, and asymmetric least squares treatment effects, for discrete, continuous, and mixed treatments. Under regularity conditions, we show that our estimator achieves rate double robustness and $\sqrt{N}$-asymptotic normality. Moreover, its asymptotic variance achieves the semiparametric efficiency bound. We further develop a statistical inference procedure based on weighted bootstrap, which avoids estimating the efficient influence/score functions. Simulation results reveal that the proposed method consistently outperforms existing alternatives, especially when the sample size is small. Applications to the 401(k) dataset and the Mother's Significant Features (MSF) dataset further illustrate the practical value of the BNNW method for estimating both average and quantile treatment effects under binary and continuous treatments, respectively.

报告人简介：张政，中国人民大学统计与大数据研究院长聘副教授、博士生导师，国家青年人才计划入选者。担任中国现场统计研究会统计交叉科学研究分会常务理事、因果推断分会理事。长期从事统计因果推断的理论、方法和应用研究，在JRSS-B，JOE, JMLR, Quantitative Economics等期刊发表论文20篇。