plot.leverage.ppm {spatstat.model}

Plot Leverage Function

Description

Generate a pixel image plot, or a contour plot, or a perspective plot, of a leverage function that has been computed by leverage.ppm.

Usage

 ## S3 method for class 'leverage.ppm'
plot(x, ...,
                             what=c("smooth", "nearest", "exact"),
                             showcut=TRUE,
                             args.cut=list(drawlabels=FALSE), 
                             multiplot=TRUE)

 ## S3 method for class 'leverage.ppm'
contour(x, ...,
                             what=c("smooth", "nearest"),
                             showcut=TRUE,
                             args.cut=list(col=3, lwd=3, drawlabels=FALSE), 
                             multiplot=TRUE)

 ## S3 method for class 'leverage.ppm'
persp(x, ...,
               what=c("smooth", "nearest"),
               main, zlab="leverage")

Arguments

x

Leverage function (object of class "leverage.ppm") computed by leverage.ppm.

...

Arguments passed to plot.im or contour.im or persp.im controlling the plot.

what

Character string (partially matched) specifying the values to be plotted. See Details.

showcut

Logical. If TRUE, a contour line is plotted at the level equal to the theoretical mean of the leverage.

args.cut

Optional list of arguments passed to contour.default to control the plotting of the contour line for the mean leverage.

multiplot

Logical value indicating whether it is permissible to display several plot panels.

main

Optional main title. A character string or character vector.

zlab

Label for the z axis. A character string.

Details

These functions are the plot, contour and persp methods for objects of class "leverage.ppm". Such objects are computed by the command leverage.ppm.

The plot method displays the leverage function as a colour pixel image using plot.im, and draws a single contour line at the mean leverage value using contour.default. Use the argument clipwin to restrict the plot to a subset of the full data.

The contour method displays the leverage function as a contour plot, and also draws a single contour line at the mean leverage value, using contour.im.

The persp method displays the leverage function as a surface in perspective view, using persp.im.

Since the exact values of leverage are computed only at a finite set of quadrature locations, there are several options for these plots:

what="smooth":

(the default) an image plot showing a smooth function, obtained by applying kernel smoothing to the exact leverage values;

what="nearest":

an image plot showing a piecewise-constant function, obtained by taking the exact leverage value at the nearest quadrature point;

what="exact":

a symbol plot showing the exact values of leverage as circles, centred at the quadrature points, with diameters proportional to leverage.

The pixel images are already contained in the object x and were computed by leverage.ppm; the resolution of these images is controlled by arguments to leverage.ppm.

Value

Same as for plot.im, contour.im and persp.im respectively.

Author(s)

Adrian Baddeley [e-mail], Rolf Turner [e-mail] and Ege Rubak [e-mail].

References

Baddeley, A., Chang, Y.M. and Song, Y. (2013) Leverage and influence diagnostics for spatial point process models. Scandinavian Journal of Statistics 40, 86–104.

See Also

leverage.ppm.

Examples

if(offline <- !interactive()) op <- spatstat.options(npixel=32, ndummy.min=16)

   X <- rpoispp(function(x,y) { exp(3+3*x) })
   fit <- ppm(X ~x+y)
   lef <- leverage(fit)
   plot(lef)
   contour(lef)
   persp(lef)

if(offline) spatstat.options(op)