cv.bh3 | R Documentation |
Several alterations to cv.bh()
were necessary to ensure that
update()
works in the functions compare_ssnet()
. Many
arguments and functionality are the same as cv.bh()
. See
cv.bh
for details. An addition in this version is
also that for binary outcomes classification and observed accuracy,
sensitivity, specificity, and positive and negative predictive values can
be output as well as the orginally included measures.
cv.bh3(
object,
nfolds = 10,
foldid = NULL,
fold.seed = NULL,
ncv = 1,
verbose = TRUE,
classify = FALSE,
classify.rule = 0.5
)
object |
a fitted object. |
nfolds |
number of folds(groups) into which the data should be split to estimate the cross-validation prediction error. default is 10. Although nfolds can be as large as the sample size (leave-one-out CV), it is not recommended for large datasets. |
foldid |
an optional vector (if ncv = 1) or matrix (if ncv > 1) of values between 1 and nfolds identifying what fold each observation is in.
If supplied, nfolds can be missing.If |
fold.seed |
An integer that sets the seed for generating folds. |
ncv |
repeated number of cross-validation. |
verbose |
logical. If |
classify |
Logical. When |
classify.rule |
A value between 0 and 1. For a given predicted value
from a logistic regression, if the value is above |
The package pROC
will not calculate the AUC when a fold does
does not have at least one observation of each level. This can largely be
avoided by selecting the number of folds so that such circumstances are
rare. When such does occur, the current result is to assign AUC <- NA.
Note that during cross validation, the initialization values for the
algorithm to re-fit the model are the initial estimates for the
object
. This follows cv.bh
.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.