Probability residual of ordinal logistic regreession {MXM}

Probability residual of ordinal logistic regreession

Description

Probability residual of ordinal logistic regreession.

Usage

ord.resid(y, est)

Arguments

y

An ordered factor variable or a numerical vector.

est

A matrix with the fitted values of an ordinal logistic regression model.

Details

The probability residual of an ordinal logistic regression model is calculated (Li and Shepherd, 2012). It is a vector, irrespective of how many categories there are.

Value

A vector with the probability residuals.

Author(s)

Michail Tsagris

R implementation and documentation: Michail Tsagris [e-mail]

References

Li C., & Shepherd B. E. (2012). A new residual for ordinal outcomes. Biometrika, 99(2): 473–480.

See Also

testIndOrdinal, ordinal.reg

Examples

library(MASS)
y <- factor( rbinom(400, 3, 0.6), ordered = TRUE )
x <- rnorm(400)
mod <- MASS::polr(y ~ x) 
res <- ord.resid(y, mod$fitted.values)