A0359
Title: Nonparametric multiple-output center-outward quantile regression
Authors: Alberto Gonzalez Sanz - Columbia University (United States) [presenting]
Abstract: The problem of nonparametric multiple-output quantile regression is addressed based on the novel concept of multivariate center-outward quantiles introduced in past studies. To obtain conditional quantile regions and contours, the conditional center-outward quantiles are defined. A new cyclically monotone interpolation, with non-necessarily constant weights, is proposed to define them. This method is completely nonparametric and produces interpretable empirical regions/contours that converge in probability with its population counterpart. Some real and synthetic examples are included, showing the adaptation and applicability of the method, in particular its ability to catch the heteroskedasticity and the trend of the data.
