Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

logitFD: an R package for functional principal component logit regression

Manuel Escabias, Ana M. Aguilera, Christian Acal

TL;DR

The logitFD package introduced here provides a toolbox for the fit of these models by implementing the different proposed solutions and by generalizing the model proposed in 2004 to the case of several functional and non-functional predictors.

Abstract

The functional logit regression model was proposed by Escabias et al. (2004) with the objective of modeling a scalar binary response variable from a functional predictor. The model estimation proposed in that case was performed in a subspace of L2(T) of squared integrable functions of finite dimension, generated by a finite set of basis functions. For that estimation it was assumed that the curves of the functional predictor and the functional parameter of the model belong to the same finite subspace. The estimation so obtained was affected by high multicollinearity problems and the solution given to these problems was based on different functional principal component analysis. The logitFD package introduced here provides a toolbox for the fit of these models by implementing the different proposed solutions and by generalizing the model proposed in 2004 to the case of several functional and non-functional predictors. The performance of the functions is illustrated by using data sets of functional data included in the fda.usc package from R-CRAN.

logitFD: an R package for functional principal component logit regression

TL;DR

The logitFD package introduced here provides a toolbox for the fit of these models by implementing the different proposed solutions and by generalizing the model proposed in 2004 to the case of several functional and non-functional predictors.

Abstract

The functional logit regression model was proposed by Escabias et al. (2004) with the objective of modeling a scalar binary response variable from a functional predictor. The model estimation proposed in that case was performed in a subspace of L2(T) of squared integrable functions of finite dimension, generated by a finite set of basis functions. For that estimation it was assumed that the curves of the functional predictor and the functional parameter of the model belong to the same finite subspace. The estimation so obtained was affected by high multicollinearity problems and the solution given to these problems was based on different functional principal component analysis. The logitFD package introduced here provides a toolbox for the fit of these models by implementing the different proposed solutions and by generalizing the model proposed in 2004 to the case of several functional and non-functional predictors. The performance of the functions is illustrated by using data sets of functional data included in the fda.usc package from R-CRAN.
Paper Structure (26 sections, 25 equations, 5 figures)

This paper contains 26 sections, 25 equations, 5 figures.

Table of Contents

  1. logitFD: an R package for functional principal component logit regression
  2. Introduction
  3. https://CRAN.R-project.org/package=logitFD package: general statements
  4. Functional data analysis
  5. Functional logit regression model
  6. Extended functional logit model: several functional and nonfunctional predictors
  7. Ordinary functional principal components logit regression
  8. glm.fit object of Fit1:
  9. Intercept object of Fit1:
  10. betalist object of Fit1:
  11. PC.variance object of Fit1:
  12. ROC.curve object of Fit1:
  13. Filtered functional principal components included in the model according to their explained variability
  14. glm.fit object of Fit2:
  15. Classification table of Fit2:
  16. ...and 11 more sections

Figures (5)

  • Figure 1: Curves of mean monthly Temperature, Precipitation and Wind.
  • Figure 2: Functional parameters and Roc Curve from Fit 1 by means of logitFD.pc() function.
  • Figure 3: Functional parameters and Roc Curve from Fit 2 by means of logitFD.fpc() function.
  • Figure 4: Functional parameters and Roc Curve from Fits 3 and 4 by means of logitFD.pc.step() and logitFD.fpc.step() functions respectively.
  • Figure 5: Steps for Functional Principal Componets Logit Regression fit in its different situations considered in https://CRAN.R-project.org/package=logitFD package.