plTA.polr: Plot Residuals against Fitted Values for polr objects

In regr0: Building regression models

Description

Displays the "Tukey-Anscombe" plot for ordered regression models.

Usage

plTA.polr(object, colbars = grey(0.7), colref = grey(0.7), ploty = FALSE)

Arguments

object	result of a call to polr
colbars	color to be used for plotting the bars representing the residuals
colref	color for reference line
ploty	if TRUE, the latent response will be plotted instead of the residuals

Details

The plot intends to show the residuals for the latent variable Z that determines the response values Y. Since the precise residuals of the latent variable cannot be known, the plot shows their conditional distribution, given the observed response values. This conditional distribution is characerized by its median (by -) and the two other quartiles (by vertical bars).

Value

none.

Author(s)

Werner A. Stahel, ETH Zurich

References

See http://stat.ethz.ch/~stahel/regression

See Also

plot.regr, residuals.polr

Examples

require(MASS) # polr() and 'housing' data
house.plr <- polr(Sat ~ Infl + Type + Cont, weights = Freq, data = housing)
plTA.polr(house.plr)

regr0 documentation built on May 2, 2019, 4:52 p.m.