In theory, two-stage least squares (TSLS) identifies a weighted average of covariate-specific local average treatment effects (LATEs) from a saturated specification, without making parametric assumptions on how available covariates enter the model. In practice, TSLS is severely biased as saturation leads to a large number of control dummies and an equally large number of, arguably weak, instruments. This paper derives asymptotically valid tests and confidence intervals for the weighted average of LATEs that is targeted, yet missed by saturated TSLS. The proposed inference procedure is robust to unobserved treatment effect heterogeneity, covariates with rich support, and weak identification. We find LATEs statistically significantly different from zero in applications in criminology, finance, health, and education.
Causal inference in two-sided randomization designs: factorial regression, two-way clustering, and covariate adjustment
Qiuzhen College, Tsinghua University(清华大学邱士书院);HSBC Business School, Peking University(北京大学汇丰商学院);Adobe Research(Adobe研究院);Department of Statistics, University of California, Berkeley(加州大学伯克利分校统计系)
Inference under Covariate-Adaptive Randomization with Multiple Treatments
本文研究了多个治疗方案下协变量自适应随机化随机对照试验的推断问题。更具体地说,我们研究了一种或多种治疗相对于其他治疗或对照的平均效果的推断。如Bugni et al.(2018)所述,协变量自适应随机化是指首先根据基线协变量进行分层,然后分配治疗状态以实现各分层内平衡的随机化方案。与Bugni et al.(2018)相比,我们不仅允许多种治疗,而且还允许分配给每种治疗的单位比例在层间变化。我们首先研究来自完全饱和线性回归的估计量的性质,即,每种治疗的指标与每一层的指标之间所有相互作用的结果的线性回归。我们表明,这些估计的基础上使用通常的异方差一致估计的渐近方差的测试是无效的,另一方面,测试的基础上,这些估计和适当的估计,我们提供的渐近方差是准确的。对于分配给每个处理的目标单位比例在不同层之间没有变化的特殊情况,我们还考虑了基于从具有层固定效应的线性回归中得出的估计量的检验,即,每种治疗的指标和每一层的指标的结果的线性回归。我们表明,这些估计的基础上使用通常的异方差一致估计的渐近方差的测试是保守的,但测试的基础上,这些估计和适当的估计,我们提供的渐近方差是准确的。模拟研究说明了我们的理论结果的实际意义。