Regression#

Module with classes to perform regression of functional data.

Linear regression#

A linear regression model is one in which the response variable can be expressed as a linear combination of the covariates (which could be multivariate or functional). The following linear models are available in scikit-fda:

skfda.ml.regression.LinearRegression(*[, ...])

Linear regression with multivariate and functional response.

skfda.ml.regression.HistoricalLinearRegression(*, ...)

Historical functional linear regression.

Nearest Neighbors#

This module contains nearest neighbors estimators to perform regression. In the examples Neighbors Scalar Regression and Neighbors Functional Regression it is explained the basic usage of these estimators.

skfda.ml.regression.KNeighborsRegressor()

Regression based on k-nearest neighbors.

skfda.ml.regression.RadiusNeighborsRegressor()

Regression based on neighbors within a fixed radius.

Kernel regression#

This module includes the implementation of Kernel Regression for FData with a scalar as a response variable. It is a non-parametric technique that uses HatMatrix object.

skfda.ml.regression.KernelRegression(*[, ...])

Kernel regression with scalar response.

FPCA regression#

This module includes the implementation of FPCA (Functional Principal Component Analysis) regression for FData with scalar response. FPCA regression consists of two steps. Firstly, the principal component basis is created. Then a linear regression is fitted using the coefficients of the functions in said basis.

skfda.ml.regression.FPCARegression([...])

Regression using Functional Principal Components Analysis.

FPLS regression#

This module includes the implementation of FPLS (Functional Partial Least Squares) regression. This implementation accepts either functional or multivariate data as the regressor and the response. FPLS regression consists on performing the FPLS dimensionality reduction algorithm but using a regression deflation strategy.

skfda.ml.regression.FPLSRegression([...])

Regression using Functional Partial Least Squares.