Randomized experiments from non-random selection in US House elections
Randomized experiments from non-random selection in US House elections
复制标题
DOI：
10.1016/j.jeconom.2007.05.004
复制
发表时间：
2008-02-01
影响因子：
6.3
通讯作者：
Lee, David S.
中科院分区：
文献类型：
--
作者：
Lee, David S.
本文建立了一个相对较弱的条件下，因果推论从回归不连续（RD）分析可以作为可信的随机试验，因此，在此条件下，RD设计的有效性可以通过检查是否有一个不连续的任何预先确定的（或“基线”）的变量在RD阈值进行测试。具体来说，考虑一个标准的治疗评估问题，其中当且仅当V > nu（0）时，治疗被分配给个体，但其中vo是已知的阈值，V是可观察的。V可以取决于个人的特征和选择，但也有一个随机的机会元素：对于每个人来说，存在一个定义良好的概率分布V的密度函数-允许在整个人口中任意不同-假设是连续的。正式确定的是，这里的治疗状态在V = nu（0）的局部邻域中与随机化一样好。这些想法在对美国众议院选举的分析中得到了说明，最终计票结果的内在不确定性是合理的，这意味着获胜的政党基本上是在以微弱优势决定的选举中随机产生的。证据与这一预测是一致的，然后用于产生“接近实验”的因果估计的选举优势在职。(c)2007 Elsevier B. V.保留所有权利。
This paper establishes the relatively weak conditions under which causal inferences from a regression-discontinuity (RD) analysis can be as credible as those from a randomized experiment, and hence under which the validity of the RD design can be tested by examining whether or not there is a discontinuity in any pre-determined (or "baseline") variables at the RD threshold. Specifically, consider a standard treatment evaluation problem in which treatment is assigned to an individual if and only if V > nu(0), but where vo is a known threshold, and V is observable. V can depend on the individual's characteristics and choices, but there is also a random chance element: for each individual, there exists a well-defined probability distribution for V. The density function-allowed to differ arbitrarily across the population-is assumed to be continuous. It is formally established that treatment status here is as good as randomized in a local neighborhood of V = nu(0). These ideas are illustrated in an analysis of U.S. House elections, where the inherent uncertainty in the final vote count is plausible, which would imply that the party that wins is essentially randomized among elections decided by a narrow margin. The evidence is consistent with this prediction, which is then used to generate "near-experimental" causal estimates of the electoral advantage to incumbency. (c) 2007 Elsevier B.V. All rights reserved.