plot.dynaTree: Plotting Predictive Distributions of Dynamic Tree models

Description

Plotting predictive distributions constructed from dynamic tree (regression) models for 1-d data -- provided primarily for use in our 1-d examples and for illustrative purposes

Usage

# S3 method for dynaTree
plot(x, proj = 1, add = FALSE, ylim = NULL,
              col = 2, lwd = 1, ptype = c("each", "mean"), ...)

Value

The only output of this function is a pretty plot

Arguments

x: a "dynaTree"-class object built by dynaTree
add: a scalar logical indicating if the lines/points should be “added” to an existing plot
proj: when ncol(x$X) > 1 this argument can be used to plot a 1-d projection by specifying which column of x$X should be used to make the plot
ylim: user-specified y-axis limits values; see plot
col: user-specified color value; see plot
lwd: user-specified line-width value; see plot
ptype: type of plot used to visualize several predictive samples obtained from dynaTrees: "each" shows each surface with its own set of three lines, and "mean" shows the three lines obtained by averaging
...: other arguments to the generic plot method

Author

Robert B. Gramacy rbg@vt.edu,
Matt Taddy and Christoforos Anagnostopoulos

Details

This plotting function only handles the predictive output from 1-dimensional regression dynaTree models as obtained by first calling dynaTree and then predict.dynaTree on the resulting output at new XX locations. It is provided to help make the illustration of our 1-d examples easier and to serve as an aid in a user's development of custom plotting functions in higher dimensions

References

Taddy, M.A., Gramacy, R.B., and Polson, N. (2011). “Dynamic trees for learning and design” Journal of the American Statistical Association, 106(493), pp. 109-123; arXiv:0912.1586

https://bobby.gramacy.com/r_packages/dynaTree/

Examples

Run this code

## see dynaTree, dynaTrees and update.dynaTree for examples
## which use this plot function

Run the code above in your browser using DataLab