verification: discrimination.plot – R documentation

Become an expert in R — Interactive courses, Cheat Sheets, certificates and more!

discrimination.plot

Discrimination plot

Description

This function creates a plot of discrimination plots (overlay histograms). In the context of verification, this is often used to compare the distribution of event and no-event forecasts. This may be useful in comparing any set of observations. By default, boxplots of groups appear as upper marginal plots. These may be surpressed.

Usage

discrimination.plot(group.id, value, breaks = 11, main =
"Discrimination Plot", xlim = NULL, ylim = NULL,  legend =
FALSE, leg.txt = paste("Model", sort(unique(group.id)) ),   marginal = TRUE, cols =
seq(2, length(unique(group.id)) + 1), xlab = "Forecast",  ... )

Arguments

`group.id`	A vector identifying groups. A histogram is created for each unique value.
`value`	A vector of values corresponding to the group.id vector used to create the histograms
`breaks`	Number of breaks in the x-axis of the histogram. The range of values is taken to be the range of prediction values.
`main`	Title for plot.
`xlim`	Range of histogram - x axis - main plot coordinates.
`ylim`	Range of histogram - y axis - main plot coordinates.
`legend`	Should there be a legend? Default = FALSE
`leg.txt`	Legend text. If FALSE or if a marginal plot is created, no legend is added.
`cols`	A vector showing the colors to be used in the histograms and in the marginal boxplots
`marginal`	Should a boxplots be placed in the top margin? Defaults to TRUE
`xlab`	Label of the x-axis on the main plot.
`...`	Additional plotting options.

Author(s)

Matt Pocernich

Examples

#  A sample forecast.  

data(disc.dat)
discrimination.plot(disc.dat$group.id, disc.dat$frcst, main = "Default  Plot")

discrimination.plot(disc.dat$group.id, disc.dat$frcst, main = "New Labels", cex = 1.2,
leg.txt = c("Low", "Med", "High" ) )

discrimination.plot(disc.dat$group.id, disc.dat$frcst, main = "Without Marginal Plots ",
marginal = FALSE)

verification

Weather Forecast Verification Utilities

v1.42

GPL (>= 2)

Authors

NCAR - Research Applications Laboratory

Initial release

2015-07-10