plot1SDRule: Plot Regularization Path and One Standard Deviation Rule

In bestglm: Best Subset GLM and Regression Utilities

Description

Takes input either matrix with 2 columns or output from caret::train() and produces a plot showing the best model selected using the 1 SD rule.

Usage

plot1SDRule(ans, main = "", sub = "", xlab = "df", ylab = "EPE")

Arguments

ans	matrix or output from train
main	optional plot title
sub	optional plot subtitle
xlab	optional x-axis label
ylab	optional y-axis label

Value

tuning parameter value for best model

Author(s)

A. I. McLeod

References

Hastie, Tibsharani and Friedman, "Elements of Statistical Learning".

See Also

oneSDRule

Examples

CV<-c(1.4637799,0.7036285,0.6242480,0.6069406,0.6006877,0.6005472,0.5707958,
      0.5907897,0.5895489)
CVsd<-c(0.24878992,0.14160499,0.08714908,0.11376041,0.08522291,
 0.11897327,0.07960879,0.09235052,0.12860983)
CVout <- matrix(c(CV,CVsd), ncol=2)
oneSDRule(CVout)

Example output

Loading required package: leaps
[1] 3

bestglm documentation built on March 26, 2020, 7:25 p.m.