CMStatistics 2021: Start Registration
View Submission - CMStatistics
Title: A general approach for nonparametric regression with circular predictors Authors:  Maria Alonso-Pena - Universidade de Santiago de Compostela (Spain) [presenting]
Irene Gijbels - KU Leuven (Belgium)
Rosa Crujeiras - University of Santiago de Compostela (Spain)
Abstract: Circular data are observations that can be represented on the circumference of the unit circle, such as angles and directions, and are found in many different fields, for example, biology, meteorology or even psychology. Moreover, circular observations can be found jointly with other variables, and several real-life problems involve the estimation of a regression function when the predictor (or one of the predictors) is of a circular nature. We introduce a broad approach that allows the nonparametric estimation of regression functions with a circular predictor and a general response, which can be either a continuous or a discrete variable. We will present the estimation procedure, which consists of the maximization of the circular kernel weighted log-likelihood, and address the problem of the selection of the smoothing parameter. The finite sample performance of the estimation method will be explored and some real data applications will be presented.