predict.glmnetcr: AIC, BIC, Predicted Class, and Fitted Probabilities for All...

In glmnetcr: Fit a Penalized Constrained Continuation Ratio Model for Predicting an Ordinal Response

Description

Returns the AIC, BIC, predicted class, and the fitted probabilities for the K classes for all steps along the path.

Usage

## S3 method for class 'glmnetcr'
predict(object, newx = NULL, ...)

Arguments

object	a glmnetcr object
newx	a data matrix representing the predictor variables, if missing defaults to original data used in fitting the model
...	additional optional arguments

Value

AIC	a numeric vector representing the AIC for each step
BIC	a numeric vector representing the BIC for each step
class	a matrix with n rows indicating the predicted class for each observation in newx for each stepp
probs	an array where for each step, a matrix of n rows and K columns indicates the fitted class probabilities for each observation and class

Author(s)

Kellie J. Archer, [e-mail]

See Also

See Also as glmnetcr, fitted.glmnetcr

Examples

data(diabetes)
x <- diabetes[, 2:dim(diabetes)[2]]
y <- diabetes$y
glmnet.fit <- glmnetcr(x, y)
result <- predict(glmnet.fit)
names(result)

Example output

Loading required package: glmnet
Loading required package: Matrix
Loading required package: foreach
Loaded glmnet 2.0-12

Warning message:
from glmnet Fortran code (error code -26); Convergence for 26th lambda value not reached after maxit=100 iterations; solutions for larger lambdas returned 
[1] "BIC"   "AIC"   "class" "probs"

glmnetcr documentation built on July 8, 2020, 6:21 p.m.