PCV_grid: Partitioned cross-validation

In hoangtt1989/NOIS: Non-parametric Outlier Identification and Smoothing

Description

Evaluate partitioned cross-validation (PCV) for the Nadaraya-Watson estimator over a grid

Usage

PCV_grid(x, y, hgrid = seq(from = 0.05, to = 3, length.out = 100),
  g = floor(length(x)/5), weight_probs = c(0.05, 0.95))

Arguments

x	'x' values.
y	'y' values.
hgrid	The grid of test bandwidths.
g	The number of points used in each training subgroup. This is the most sensitive parameter.
weight_probs	Specification for what percentage of points are used in the weight function.

Value

The minimum bandwidth h, the minimum CV error and the grid of test bandwidths.

See Also

Other NOIS CV functions: LOOCV_grid, MCV_grid, NOIS_fit

hoangtt1989/NOIS documentation built on May 20, 2019, 2:08 p.m.