M. Alonso Pena, G. Claeskens, I. Gijbels
In this talk we introduce a regression model for a circular response variable and multiple covariates. To flexibly estimate the regression function while not suffering from the curse of dimensionality, we consider a single-index approach. An algorithm is presented to compute estimators for both the parametric and nonparametric parts of the model, and some asymptotic results are given for both estimators. Finally, some simulation experiments are presented and a real data illustration on wave direction modelling is provided.
Palabras clave: Circular data, Directional data, kernel smoothing, regression, multiple covariates
Programado
Estimación no paramétrica
12 de junio de 2025 19:00
Sala de prensa (MR 13)