Constrained Groupwise Additive Index Models

Fits constrained groupwise additive index models and provides functions for inference and interpretation of these models. The method is described in Masselot, Chebana, Campagna, Lavigne, Ouarda, Gosselin (2022) "Constrained groupwise additive index models" .

Constrained Groupwise Additive Index Models

Description

The cgaim package allows fitting Groupwise Additive Index Models with constraints (CGAIM) on both the ridge functions and indices coefficients. Methods to plot the ridge functions, predict new data and compute confidence intervals are also included in the package.

The model and fitting are described in:

Pierre Masselot, Fateh Chebana, Céline Campagna, Éric Lavigne, Taha B M J Ouarda, Pierre Gosselin (2022). Constrained groupwise additive index models. Biostatistics, 24(04), 1066–1084. https://doi.org/10.1093/biostatistics/kxac023.

Installation

The package is available for installation from the usual CRAN repository. Alternatively, to install the development version:

1. In R, install the package directly from Github using the command (the package devtools is required):

> library(devtools)
> install_github("PierreMasselot/cgaim")

1. The package can then be loaded as usual: library(cgaim).
2. Help can be accessed from R with ?cgaim.

Functions

The main function of the package is the eponymous cgaim that fits the model. Then, the print.cgaim method allows displaying the results. The confint.cgaim and vcov.cgaim methods allow computing confidence intervals and variance-covariance matrix using various approaches. Ridge functions can be displayed using the plot.cgaim method, with the possibility to add confidence intervals. Finally, predict.cgaim allows computing the indices and perform prediction on new observations. See the help of each function.