FKSUM: Fast Kernel Sums
Implements the method of Hofmeyr, D.P. (2021) <doi: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
The package is accompanied by an instructive paper in the Journal of Statistical Software <doi:10.18637/jss.v101.i03>.
Please use the canonical form
to link to this page.