nested.compare: Compare cross validation fit performances from a...

In glmnetr: Nested Cross Validation for the Relaxed Lasso and Other Machine Learning Models

Compare cross validation fit performances from a nested.glmnetr output.

Description

Compare cross-validation model fits in terms of average performances from the nested cross validation fits. In general the standard deviations for the performance measures evaluated on the leave-out samples may be biased. While the standard deviations of the paired within fold differences of performances intuitively might be less biased this has not been shown. See the package vignettes for more discussion.

Usage

nested.compare(object, type = "devrat", digits = 4, pow = 1)

Arguments

object	A nested.glmnetr output object.
type	determines what type of nested cross validation performance measures are compared. Possible values are "devrat" to compare the deviance ratios, i.e. the fractional reduction in deviance relative to the null model deviance, "agree" to compare agreement, "lincal" to compare the linear calibration slope coefficients, "intcal" to compare the linear calibration intercept coefficients, from the nested cross validation.
digits	digits for printing of z-scores, p-values, etc. with default of 4
pow	the power to which the average of correlations is to be raised. Only applies to the "gaussian" model. Default is 2 to yield R-square but can be on to show correlations. pow is ignored for the family of "cox" and "binomial".

Value

A printout to the R console.

See Also

nested.cis , summary.nested.glmnetr , nested.glmnetr

Examples

sim.data=glmnetr.simdata(nrows=1000, ncols=100, beta=NULL)
xs=sim.data$xs 
y_=sim.data$yt
event=sim.data$event
# for this example we use a small number for folds_n to shorten run time 
fit3 = nested.glmnetr(xs, NULL, y_, event, family="cox", folds_n=3) 
nested.compare(fit3)

glmnetr documentation built on April 3, 2025, 6:45 p.m.