nsgp (version 1.0.5)

plot.gppack: Plots several GP's simultaneously

Description

Plots the GP's corresponding to the control and case data, as well as the null model. Visualizes the log likelihood ratios between the null and individual models. Several boolean parameters for modifying the plot. By default plots the data, posterior mean and 95% interval for CASE and CONTROL.

Usage

"plot"(x, y = NULL, plotdata = TRUE, plotmeans = TRUE, plotcovs = TRUE, plotnoises = FALSE, plotnull = FALSE, plotratios = "emll", thr = 1, samples = 0, sigma = 2, title = NULL, legend = FALSE, plotgradient = TRUE, ...)

Arguments

x

the gppack-object

y

placeholder variable

plotdata

plot the data (default)

plotmeans

plot the GP mean (default)

plotcovs

plot the GP covariances (default)

plotnoises

plot the observational noise (default)

plotnull

plots also the null model

plotratios

plots the ratios, choices are emll, mll, npc, pc

thr

ratio threshold

samples

plot N samples from the GP

sigma

variance level to plot

title

plot title

legend

plot legend

plotgradient

use gradient graphics

...

...

Details

The threshold thr is the logarithmic likelihood ratio between null and control+case models. The default value 1 hence corresponds to a likelihood ratio of 2.72.

Examples

# read toy data
data(toydata)

## Not run: can take several minutes
# # perform two-sample regression
# res = gpr2sample(toydata$ctrl$x, toydata$ctrl$y, toydata$case$x, toydata$case$y, seq(0,22,0.1))
# 
# # pre-computed model for toydata
# data(toygps)
# res = toygps
# 
# # basic plot
# plot(res)
# 
# # plot also the null model, don't plot data, means or noise
# plot(res, plotnull=TRUE, plotdata=FALSE, plotmeans=FALSE, plotnoise=FALSE)## End(Not run)