The joint asymptotic normality of the conditional quantiles



Abstract: Let (X,Y)  be a two dimensional random variable with a joint distribution function F(X,Y). This paper studies the kernel estimation of the conditional quantiles of for a given value of  based on a random sample from the above distribution, which was proposed by [12].

 In this paper, the joint asymptotic normality of the conditional quantile estimated at a finite number of distinct points is established under some regularity conditions. Moreover, the performance of the conditional quantile estimation in constructing prediction intervals is tested through two applications. The first application deals with simulated data set and the second deals with real life data set.

DOI Code: 10.1285/i20705948v4n1p39

Keywords: Kernel estimation, conditional distribution, conditional quantile, multivariate distribution

Full Text: PDF

Creative Commons License
This work is licensed under a Creative Commons Attribuzione - Non commerciale - Non opere derivate 3.0 Italia License.