非参数方法估计分位数模型的研究综述

The Summary of Nonparametric Quantile Estimation

目前国内对于线性分位数回归模型的研究及其应用正在火热进行中,而非参数计量方法由于其优越性也受到越来越多人的重视,但对于两种方法结合应用的相关研究才刚刚开始.对于目前国内外已有的以及正在发展中的非参数方法估计分位数回归模型做了一个综述,包括理论综述及实证应用综述,其中重点介绍了此方面研究的前沿理论:Li and Racine关于非参数估计条件累积分布函数(CDF)和分位函数的研究.通过介绍,希望对于我国非参数估计分位数回归模型方法的研究和应用有所裨益.

At present, the research and the applications of linear quantile regression model are very popular. Because of its superiority, people have paid more and more attention to Nonparametric Method at the same time. However, the combination of two methods for the application has only just begun In this paper, an overview which includes both at home and abroad for the existing as well as the development of the use of non-parametric methods to estimate quantile regression model is introduced. We introduce both theory and empirical applications, and then we focus on the forefront of theory： Nonparametric Estimation of Conditional CDF and Quantile Functions with Mixed Categorical and Continuous Data （the paper of Li and Racine）. Through the introduction of this article, I hope this method can be extensive popularized.