标题： 基于半参数单指标模型的异质处理效应统计推断 显示英文标题

标题： Statistical inference of heterogeneous treatment effects using semiparametric single-index model

摘要： 近年来，随着科技的迅速发展，异质处理效应已成为统计学、计量经济学和社会学中的研究热点。 本文通过双重稳健估计的半参数单指标模型来研究HTE。 与传统方法不同，我们既不对单指标模型中的连接函数施加有界性约束，也不限制其支撑范围。 通过采用筛法近似连接函数，我们实现了对连接函数和指标参数的同时估计。 我们的研究不仅建立了所提出估计量的渐近性质，还通过全面的模拟研究系统评估了其小样本性能。 数值结果表明，我们的方法显著优于其他常用的竞争估计器。 此外，我们将所提出的方法应用于国家健康与营养检查调查数据集，以评估参与学校午餐计划对体重指数的影响。 显示英文摘要

摘要： In recent years, with the rapid development of science and technology, heterogeneous treatment effects have emerged as a focal research topic in statistics, econometrics, and sociology. This paper investigates HTE through semiparametric single-index models based on doubly robust estimation. Departing from conventional approaches, we neither impose boundedness constraints on the link function in single-index models nor restrict its support range. By employing the sieve method to approximate the link function, we achieve simultaneous estimation of both the link function and index parameters. Our study not only establishes the asymptotic properties of the proposed estimator but also systematically evaluates its finite-sample performance through comprehensive simulation studies. Numerical results demonstrate that our method significantly outperforms other commonly used competing estimators. Furthermore, we apply the proposed approach to the National Health and Nutrition Examination Survey dataset to assess the impact of participation in school lunch programs on body mass index.