期刊名称：Journal of Statistical and Econometric Methods
印刷版ISSN：2241-0384
电子版ISSN：2241-0376
出版年度：2018
卷号：7
期号：2
语种：English
出版社：Scienpress Ltd
摘要：This paper develops two new classes ofestimators measuring the distributive effects of a treatment on a population.Using imputation methods, empirical quantile and bootstrap simulations, wemanaged to define and study the properties of the two classes. The first classis Imputation Based Treatment Effect on distribution based on rank preservationassumption, basically the effect of treatment on the distribution of potentialoutcome. The second class is Imputation Based Quantile Treatment Effect which,according to this work is supposed to be the true Quantile Treatment Effectsince no rank preservation assumption is made. The second class is based on thefact that each quantile before the treatment is tracked after the treatment andthe estimator compares the same group before and after. The first class ofestimators (for example the one generated by k-Nearest Neighbors imputationmethod) performs well as classic Quantile Treatment Effect given the simulationresult. When applied to Lalonde real data set, it performs better than classicQuantile Treatment Effect and Firpo’s semi parametric estimator especially formiddle quantiles. Also, we found that there is a significant difference betweenthe two classes of estimators meaning that the bias caused by rank preservationassumption is quite significant.
