HEAPING-INDUCED BIAS IN REGRESSION-DISCONTINUITY DESIGNS
HEAPING-INDUCED BIAS IN REGRESSION-DISCONTINUITY DESIGNS
复制标题
DOI：
10.1111/ecin.12225
复制
发表时间：
2016-01-01
期刊：
影响因子：
1.8
通讯作者：
Waddell, Glen R.
中科院分区：
文献类型：
--
作者：
Barreca, Alan I.;Lindo, Jason M.;Waddell, Glen R.
本研究使用蒙特卡罗模拟证明，回归不连续性设计时，与结果相关的属性预测堆积在运行变量的估计值有偏。在表明我们通常的诊断可能不适合识别这种类型的问题后，我们提供了替代方案，然后讨论了不同方法来解决偏差的有用性。然后，我们在多个非模拟环境中考虑这些问题。(JEL C21、C14、I12）
This study uses Monte Carlo simulations to demonstrate that regression-discontinuity designs arrive at biased estimates when attributes related to outcomes predict heaping in the running variable. After showing that our usual diagnostics may not be well suited to identifying this type of problem, we provide alternatives, and then discuss the usefulness of different approaches to addressing the bias. We then consider these issues in multiple non-simulated environments. (JEL C21, C14, I12)