fsemipar: an R package for estimation, variable selection and prediction for functional semiparametric models
This package is devoted to estimation or simultaneous estimation and variable selection of several functional semiparametric models with a scalar response. That includes the functional single-index model, the semi-functional partial linear model, and the semi-functional partial linear single-index model. It also contains algorithms for addressing estimation and variable selection in the linear model, the multi-functional partial linear model, and the multi-functional partial linear single-index model when the scalar covariates with linear effects come from the discretization of a curve. In addition, the package has routines for kernel- and kNN-based estimation with Nadaraya-Watson weights of models with a nonparametric component. It also contains functions to compute predictions from all the considered models and estimation procedures.
Palabras clave: R package Functional data analysis Variable selection Semiparametrics Multi-functional covariates Functional single-index model