Inference With Difference-in-Differences With a Small Number of Groups A Review, Simulation Study, and Empirical Application Using SHARE Data
Inference With Difference-in-Differences With a Small Number of Groups A Review, Simulation Study, and Empirical Application Using SHARE Data
复制标题
DOI：
10.1097/mlr.0000000000000830
复制
发表时间：
2018-01-01
期刊：
影响因子：
3
通讯作者：
Landrum, Mary Beth
中科院分区：
文献类型：
--
作者：
Rokicki, Slawa;Cohen, Jessica;Landrum, Mary Beth
背景资料：差异中的差异（DID）估计已成为越来越受欢迎的方法来评估群体水平的政策对个人水平的结果的影响。已经提出了几种统计方法来纠正因数据聚类而产生的模型误差的组内相关性。鲜为人知的是，这些校正如何执行与经常观察到的健康研究中使用纵向data.Methods小组的数量很少：首先，我们回顾了最常用的建模解决方案，在DID估计面板数据，包括广义估计方程（GEE），排列检验，聚类标准误（CSE），野生集群引导，和聚合。其次，我们比较经验的覆盖率和功率这些方法使用Monte Carlo模拟研究的情况下，我们不同程度的误差相关性，组大小的平衡，和治疗组的比例。第三，我们提供了一个实证的例子，使用调查的健康，老龄化和退休在European.Results：当组的数量是小的，CSE系统性地向下偏置的情况下，当数据是不平衡的，或者当有一个低比例的治疗组。这可能导致过度拒绝空值，即使数据由多达50个组组成。聚合，置换测试，偏差调整GEE，野生集群引导产生的覆盖率接近名义率几乎所有的情况下，虽然GEE可能遭受低power.Conclusions：在DID估计与少数群体，分析使用聚合，置换测试，野生集群引导，或偏差调整GEE建议。
Background: Difference-in-differences (DID) estimation has become increasingly popular as an approach to evaluate the effect of a group-level policy on individual-level outcomes. Several statistical methodologies have been proposed to correct for the within-group correlation of model errors resulting from the clustering of data. Little is known about how well these corrections perform with the often small number of groups observed in health research using longitudinal data.Methods: First, we review the most commonly used modeling solutions in DID estimation for panel data, including generalized estimating equations (GEE), permutation tests, clustered standard errors (CSE), wild cluster bootstrapping, and aggregation. Second, we compare the empirical coverage rates and power of these methods using a Monte Carlo simulation study in scenarios in which we vary the degree of error correlation, the group size balance, and the proportion of treated groups. Third, we provide an empirical example using the Survey of Health, Ageing, and Retirement in Europe.Results: When the number of groups is small, CSE are systematically biased downwards in scenarios when data are unbalanced or when there is a low proportion of treated groups. This can result in over-rejection of the null even when data are composed of up to 50 groups. Aggregation, permutation tests, bias-adjusted GEE, and wild cluster bootstrap produce coverage rates close to the nominal rate for almost all scenarios, though GEE may suffer from low power.Conclusions: In DID estimation with a small number of groups, analysis using aggregation, permutation tests, wild cluster bootstrap, or bias-adjusted GEE is recommended.