cusp (version 2.3.6)

plot.cusp: Graphical Diagnostic Display of Cusp Catastrophe Data Fit

Description

This function generates diagnostic graphical displays of fits of a cusp catastrophe model to data obtained with cusp

Usage

# S3 method for cusp
plot(x, what = c("all", "bifurcation", "residual", "densities"), ...)

Value

No return value. Called for its side effect.

Arguments

x

Object returned by cusp

what

1-character string giving the type of plot desired. The following values are possible: "all" for a panel plot with all diagnostic plots, "bifurcation" for a plot of the bifurcation surface with estimated control parameter locations superimposed, "residual" for a plot of the residuals against fitted values, "densities" for a plot of density estimates conditioned on the estimated location on the bifurcation surface.

...

named arguments that are passed to lower level plotting function

Author

Raoul Grasman

Details

These diagnostic plots help to identify problems with the fitted model. In optimal cases the fitted locations in the parameter plane are dispersed over regions of qualitatively different behavior. Within each region the fitted dependent values have a density of the appropriate shape (e.g., bimodal in the bifurcation set).

References

See cusp-package

See Also

plotCuspBifurcation, plotCuspResidfitted, plotCuspDensities

Examples

set.seed(20)
x1 = runif(150)
x2 = runif(150)
z = Vectorize(rcusp)(1, 4*x1-2, 4*x2-1)
data <- data.frame(x1, x2, z)
fit <- cusp(y ~ z, alpha ~ x1+x2, beta ~ x1+x2, data)
if (FALSE) {
plot(fit)

# just densities
layout(matrix(1:4,2))
plot(fit, what="densities")
}