Learn R Programming

OSCV (version 1.0)

One-Sided Cross-Validation

Description

Functions for implementing different versions of the OSCV method in the kernel regression and density estimation frameworks. The package mainly supports the following articles: (1) Savchuk, O.Y., Hart, J.D. (2017). Fully robust one-sided cross-validation for regression functions. Computational Statistics, and (2) Savchuk, O.Y. (2017). One-sided cross-validation for nonsmooth density functions, .

Copy Link

Version

Install

install.packages('OSCV')

Monthly Downloads

163

Version

1.0

License

GPL-2

Maintainer

Olga Savchuk

Last Published

March 18th, 2017

Functions in OSCV (1.0)

h_OSCV_dens

The OSCV bandwidth in the density estimation context.
ISE_fstar

The ISE function in the kernel density estimation (KDE) context in the case when the underlying density is
h_OSCV_reg

The OSCV bandwidth in the regression context.
L_I

The family of one-sided cross-validation kernels
sample_fstar

Taking a random sample from
OSCV_LI_dens

The OSCV function based on the kernel
OSCV_reg

The OSCV function in the regression context.
OSCV_Epan_dens

The OSCV function based on
OSCV_Gauss_dens

The OSCV function based on
CV_reg

The cross-validation (CV) function in the regression context.
ASE_reg

The ASE function for the local linear estimator (LLE) in the regression context.
loclin

Computing the local linear estimate (LLE).
reg3

Nonsmooth regression function with six cusps.
fstar

Nonsmooth density function with seven cusps.
h_ASE_reg

The ASE-optimal bandwidth in the regression context.
C_smooth

The OSCV smooth rescaling constant.
H_I

The family of two-sided cross-validation kernels