plot.flamCV {flam}

Plots Cross-Validation Curve for Object of Class "flamCV"

Description

This function plots the cross-validation curve for a series of models fit using flamCV. The cross-validation error with +/-1 standard error is plotted for each value of lambda considered in the call to flamCV with a dotted vertical line indicating the chosen lambda.

Usage

## S3 method for class 'flamCV'
plot(x, showSE = T, ...)

Arguments

x

an object of class "flamCV".

showSE

a logical (TRUE or FALSE) for whether the standard errors of the curve should be plotted.

...

additional arguments to be passed. These are ignored in this function.

Author(s)

Ashley Petersen

References

Petersen, A., Witten, D., and Simon, N. (2014). Fused Lasso Additive Model. arXiv preprint arXiv:1409.5391.

See Also

flamCV

Examples

#See ?'flam-package' for a full example of how to use this package

#generate data
set.seed(1)
data <- sim.data(n = 50, scenario = 1, zerof = 0, noise = 1)

#fit model and select tuning parameters using 2-fold cross-validation
#note: use larger 'n.fold' (e.g., 10) in practice
flamCV.out <- flamCV(x = data$x, y = data$y, within1SE = TRUE, n.fold = 2)

#lambdas chosen is
flamCV.out$lambda.cv

#we can now plot the cross-validation error curve with standard errors
#vertical dotted line at lambda chosen by cross-validation
plot(flamCV.out)
#or without standard errors
plot(flamCV.out, showSE = FALSE)

## Not run: 
#can choose lambda to be value with minimum CV error
#instead of lambda with CV error within 1 standard error of the minimum
flamCV.out2 <- flamCV(x = data$x, y = data$y, within1SE = FALSE, n.fold = 2)

#contrast to chosen lambda for minimum cross-validation error
#it's a less-regularized model (i.e., lambda is smaller)
plot(flamCV.out2)

## End(Not run)