When doing impact evaluation and making causal inferences, it is important to acknowledge the heterogeneity of the treatment effects for different domains (geographic, socio-demographic, or socio-economic). If the domain of interest is small with regards to its sample size (or even zero in some cases), then the evaluator has entered the small area estimation (SAE) dilemma. Based on the modification of the Inverse Propensity Weighting estimator and the traditional small area predictors, the paper proposes a new methodology to estimate area specific average treatment effects for unplanned domains. By means of these methods we can also provide a map of policy impacts, that can help to better target the treatment group(s). We develop analytical Mean Squared Error (MSE) estimators of the proposed predictors. An extensive simulation analysis, also based on real data, shows that the proposed techniques in most cases lead to more efficient estimators.
翻译:在进行影响评估和作出因果关系推断时,必须承认不同领域(地理、社会-人口或社会经济)的治疗效果的异质性。如果所涉领域与其抽样规模(在某些情况下甚至为零)相比很小,那么评价员就进入了小面积估计(SAE)的困境。根据对倒比重估计估计器和传统的小面积预测器的修改,本文件提出一种新的方法来估计非规划领域的具体区域平均治疗效果。通过这些方法,我们还可以提供政策影响图,帮助更好地针对治疗组。我们开发了拟议预测器的分析性平均误差估计器。根据真实数据进行的广泛模拟分析表明,大多数情况下拟议的技术都导致更有效的估计器。