Internal function to conduct k-fold cross-validation for irglmreg, produces a plot,
and returns cross-validated loss values for lambda
cv.irglmreg_fit(x, y, weights, offset, lambda=NULL, balance=TRUE, cfun=4, dfun=1,
s=1.5, nfolds=10, foldid, type = c("loss", "error"), plot.it=TRUE,
se=TRUE, n.cores=2, trace=FALSE, parallel=FALSE, ...)an object of class "cv.irglmreg" is returned, which is a
list with the ingredients of the cross-validation fit.
a fitted irglmreg object for the full data.
matrix of loss values or errors with row values for lambda and column values for kth cross-validation
The mean cross-validated loss values or errors - a vector of length
length(lambda).
estimate of standard error of cv.
an optional vector of values between 1 and nfold
identifying what fold each observation is in.
a vector of lambda values
index of lambda that gives minimum cv value.
value of lambda that gives minimum cv value.
x matrix as in irglmreg.
response y as in irglmreg.
Observation weights; defaults to 1 per observation
this can be used to specify an a priori known component to be included in the linear predictor during fitting. This should be NULL or a numeric vector of length equal to the number of cases. Currently only one offset term can be included in the formula.
Optional user-supplied lambda sequence; default is
NULL, and irglmreg chooses its own sequence
for dfun=4, 5, 6 only
a number from 1 to 7, type of convex cap (concave) function
a number from 1, 4-7, type of convex downward function
nonconvex loss tuning parameter for robust regression and classification.
number of folds >=3, default is 10
an optional vector of values between 1 and nfold
identifying what fold each observation is in. If supplied,
nfold can be missing and will be ignored.
cross-validation criteria. For type="loss", loss function values and type="error" is misclassification error.
a logical value, to plot the estimated log-likelihood values if TRUE.
a logical value, to plot with standard errors.
The number of CPU cores to use. The cross-validation loop will attempt to send different CV folds off to different cores.
a logical value, print progress of cross validation or not
a logical value, parallel computing or not
Other arguments that can be passed to irglmreg.
Zhu Wang <zwang145@uthsc.edu>
The function runs irglmreg nfolds+1 times; the
first to compute the lambda sequence, and then to
compute the fit with each of the folds omitted. The error or the log-likelihood value is
accumulated, and the average value and standard deviation over the
folds is computed. Note that cv.irglmreg can be used to search for
values for alpha: it is required to call cv.irglmreg with a fixed vector foldid for different values of alpha.
Zhu Wang (2024) Unified Robust Estimation, Australian & New Zealand Journal of Statistics. 66(1):77-102.