cr.backward: Backward Continuation Ratio Restructure Function

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

Description

This is an internal function used by glmnetcr to restructure an ordinal dataset to represent the K-1 conditionally independent likelihoods needed for fitting a backrward continuation ratio model

Usage

cr.backward(x, y, weights)

Arguments

x	covariates to be included in the model
y	ordinal outcome
weights	observation weights. Can be total counts if responses are proportion matrices. Default is 1 for each observation

Value

object

Yields an object where the first column y is the ordinal outcome that has been dichotomously coded and the expanded covariate matrix for the K-1 conditionally independent likelihoods

Author(s)

Kellie J. Archer

See Also

See Also as cr.forward, glmnetcr

Examples

data(diabetes)
x <- diabetes[,2:dim(diabetes)[2]]
y <- diabetes$y
fit <- glmnetcr(x, y, method = "backward")

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