Alternative Assumptions to Identify LATE in Fuzzy Regression Discontinuity Designs
Alternative Assumptions to Identify LATE in Fuzzy Regression Discontinuity Designs
复制标题
DOI：
10.1111/obes.12249
复制
发表时间：
2018-10-01
影响因子：
2.5
通讯作者：
Dong, Yingying
中科院分区：
文献类型：
--
作者：
Dong, Yingying
存在两种替代假设来确定模糊回归不连续性 (RD) 设计中的局部平均治疗效果 (LATE)：局部独立性 (LI) 和局部平滑性 (LS)。与通常要求存在第一阶段和治疗单调性的 LATE 假设一起，这两个假设中的任何一个都足以识别 RD LATE。我讨论了这些替代假设的实际（和可测试）影响，并表明通过 LS 削弱 LI 可能具有经验相关性。然而，当 LI 确实成立时，人们可以探索一些实际意义。简要介绍了数值和经验例子。
There exist two alternative assumptions to identify local average treatment effects (LATE) in fuzzy regression discontinuity (RD) designs: local independence (LI) and local smoothness (LS). Together with the usual LATE assumptions requiring existence of a first-stage and treatment monotonicity, either of these two assumptions is sufficient to identify RD LATE. I discuss the practical (and testable) implications of these alternative assumptions, and show that weakening LI by LS might be empirically relevant. However, when LI does hold, there are some practical implications one may explore. Numerical and empirical examples are briefly presented.