KRLS: Kernel-based Regularized Least squares (KRLS)

Share:

Package implements Kernel-based Regularized Least Squares (KRLS), a machine learning method to fit multidimensional functions y=f(x) for regression and classification problems without relying on linearity or additivity assumptions. KRLS finds the best fitting function by minimizing the squared loss of a Tikhonov regularization problem, using Gaussian kernels as radial basis functions. For further details see Hainmueller and Hazlett (2014).

Author
Jens Hainmueller (Stanford) Chad Hazlett (UCLA)
Date of publication
2014-05-21 21:21:47
Maintainer
Jens Hainmueller <jhain@stanford.edu>
License
GPL (>= 2)
Version
0.3-7
URLs

View on CRAN

Man pages

fdskrls
Compute first differences with KRLS
gausskernel
Gaussian Kernel Distance Computation
krls
Kernel-based Regularized Least Squares (KRLS)
lambdasearch
Leave-one-out optimization to find lambda
looloss
Loss Function for Leave One Out Error
plot.krls
Plot method for Kernel-based Regularized Least Squares (KRLS)...
predict.krls
Predict method for Kernel-based Regularized Least Squares...
solveforc
Solve for Choice Coefficients in KRLS
summary.krls
Summary method for Kernel-based Regularized Least Squares...

Files in this package

KRLS
KRLS/NAMESPACE
KRLS/GPL-2
KRLS/R
KRLS/R/gausskernel.R
KRLS/R/looloss.R
KRLS/R/solveforc.R
KRLS/R/multdiag.R
KRLS/R/lambdasearch.r
KRLS/R/summary.krls.R
KRLS/R/predict.krls.R
KRLS/R/plot.krls.R
KRLS/R/zzz.r
KRLS/R/krls.R
KRLS/R/fdskrls.r
KRLS/MD5
KRLS/DESCRIPTION
KRLS/LICENSE.note
KRLS/man
KRLS/man/krls.Rd
KRLS/man/solveforc.Rd
KRLS/man/looloss.Rd
KRLS/man/fdskrls.Rd
KRLS/man/summary.krls.Rd
KRLS/man/lambdasearch.Rd
KRLS/man/plot.krls.Rd
KRLS/man/predict.krls.Rd
KRLS/man/gausskernel.Rd
KRLS/GPL-3