Geometrically Designed Spline (‘GeDS’) Regression is a non-parametricgeometrically motivated method for fitting variable knots splinepredictor models in one or two independent variables, in the context ofgeneralized (non-)linear models. ‘GeDS’ estimates the number andposition of the knots and the order of the spline, assuming the responsevariable has a distribution from the exponential family. A descriptionof the method can be found inKaishevet al. (2016) andDimitrova etal. (2023).
To install the stable version on R CRAN:
install.packages("GeDS")To install the latest development version:
install.packages("devtools") devtools::install_github("emilioluissaenzguillen/GeDS")This package is free and open source software, licensed underGPL-3