A. Meilan Vila, A. Panzera, A. Gottard
Circular data arise in many scientific fields where observations are recorded as directions or angles relative to a system with a fixed orientation. The circular nature of such data presents challenges when employing conventional statistical methods that are primarily designed for Euclidean data. Consequently, it is necessary to modify existing methods or develop new ones that effectively account for the characteristics of circular data.
The objective of this work is to present a procedure for estimating the parameters of a semiparametric regression model with a circular response. The asymptotic properties of the proposed estimators are discussed, and some guidelines for computing the estimates are provided. To evaluate the performance of the proposed estimators, simulations are conducted, and a real data set is employed to illustrate their behavior.
Palabras clave: Circular Data, Nonparametric Statistics, Regression Estimation
Programado
GT09.NOPAR3 Sesión Invitada. Estimación no paramétrica
9 de noviembre de 2023 11:40
HC1: Sala Canónigos 1