fancy_PPR_initialisation

<p>Computes an initial projection for PPR by replacing each covariate by the smoothed response against it. Then fit a
  standard ridge estimator to the transformed data. Not intended for general use.</p>

internal

Implements the method of Hofmeyr, D.P. (2019) <10.1109/TPAMI.2019.2930501> for fast evaluation of univariate kernel smoothers based on recursive computations.
Applications to the basic problems of density and regression function estimation are provided, as well as some projection pursuit methods
for which the objective is based on non-parametric functionals of the projected density, or conditional density of a response given projected
covariates.

David Hofmeyr

fancy_PPR_initialisation: Initialisation for PPR based on Ridge LM after GAM type smoothing

Description

Arguments