bestglm (version 0.37.3)

plot1SDRule: Plot Regularization Path and One Standard Deviation Rule

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

References

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

See Also

oneSDRule

Examples

# NOT RUN {
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)
# }