Plot Intercepts

Plots the step function corresponding to the intercepts in a orm or lrm model. This can be thought of as the link function of the covariate-adjusted empirical cumulative distribution function (actually 1 - ECDF). It is also related to q-q plots. For example, if a probit link function is an appropriate choice, and the residuals actually had a normal distribution (not needed by the semiparametric ordinal model), the step function of the intercepts would form a straight line.

plotIntercepts(fit, dots = FALSE, logt = FALSE)

Arguments

fit

an orm or lrm fit object, usually with a numeric dependent variable having many levels

dots

set to TRUE to show solid dots at the intecept values

logt

set to TRUE to use a log scale for the x-axis

Value

ggplot2 object

Author

Frank Harrell

Examples

if (FALSE) { # \dontrun{
f <- orm(y ~ x1 + x2 + x3)
plotIntercepts(f)
} # }