dismo (version 0.5-6)

gbm.plot: gbm plot

Description

Function to plot gbm response variables, with the option of adding a smooth representation of the response if requested additional options in this version allow for plotting on a common scale note too that fitted functions are now centered by subtracting their mean

Usage

gbm.plot(gbm.object, variable.no = 0, smooth = FALSE, rug = T, n.plots = length(pred.names), common.scale = T, write.title = T, y.label = "fitted function", x.label = NULL, show.contrib = T, plot.layout = c(3, 4), ...)

Arguments

gbm.object
a gbm object - could be one from gbm.step
variable.no
the var to plot - if zero then plots all
smooth
Logical. If TRUE, a smoothed version of the fitted function is added
rug
Logical. If TRUE, plot a rug of deciles is plotted
n.plots
plot the first n most important preds
common.scale
Logical. If TRUE, a common scale is used on the y axis
write.title
Logical. If TRUE, the plot gets a title
y.label
the default y-axis label
x.label
the default x-axis label
show.contrib
Logical. If TRUE, the contribution is shown on the x axis
plot.layout
define the default layout for graphs on the page
...
other arguments to pass to the plotting useful options include cex.axis, cex.lab, etc.

References

Elith, J., J.R. Leathwick and T. Hastie, 2009. A working guide to boosted regression trees. Journal of Animal Ecology 77: 802-81