KERE: Expectile Regression in Reproducing Kernel Hilbert Space

Share:

An efficient algorithm inspired by majorization-minimization principle for solving the entire solution path of a flexible nonparametric expectile regression estimator constructed in a reproducing kernel Hilbert space.

Author
Yi Yang, Teng Zhang, Hui Zou
Date of publication
2015-08-28 00:35:09
Maintainer
Yi Yang <yiyang@umn.edu>
License
GPL-2
Version
1.0.0

View on CRAN

Man pages

as.kernelMatrix
Assing kernelMatrix class to matrix objects
cv.KERE
Cross-validation for KERE
dots
Kernel Functions
KERE
Fits the regularization paths for the kernel expectile...
kernel-class
Class "kernel" "rbfkernel" "polykernel", "tanhkernel",...
kernelMatrix
Kernel Matrix functions
plot.KERE
Plot coefficients from a "KERE" object
predict.KERE
make predictions from a "KERE" object.

Files in this package

KERE
KERE/COPYING
KERE/src
KERE/src/expkern_precision.f90
KERE/src/expkern_fast.f90
KERE/NAMESPACE
KERE/R
KERE/R/predict.KERE.R
KERE/R/utilities.R
KERE/R/aobjects.R
KERE/R/kernels.R
KERE/R/cv.KERE.R
KERE/R/KERE.R
KERE/R/kernelmatrix.R
KERE/R/plot.KERE.R
KERE/MD5
KERE/DESCRIPTION
KERE/man
KERE/man/cv.KERE.Rd
KERE/man/KERE.Rd
KERE/man/as.kernelMatrix.Rd
KERE/man/dots.Rd
KERE/man/kernelMatrix.Rd
KERE/man/kernel-class.Rd
KERE/man/predict.KERE.Rd
KERE/man/plot.KERE.Rd