rgr (version 1.1.15)

gx.add.chisq: Function to Add Fences to Chi-square Plots

Description

This is an internal function used to plot fences at stated probability levels on a Chi-square plot to assist in the assessment of the plotted distribution. By default fences are plotted for the 90th, 95th and 98th percentiles of the Chi-square distribution. The function is called from gx.md.plt0, itself called from gx.md.plot that is used to display Chi-square plots generated by gx.mva, gx.mva.closed, gx.robmva, gx.robmva.closed, gx.md.gait amd gx.md.gait.closed.

Usage

gx.add.chisq(p = c(0.98, 0.95, 0.9), df = NULL, ifflip = FALSE,
	cex = 0.6)

Arguments

p

the percentiles of the Chi-square distribution for the fences to be displayed, by default the 90th, 95th and 98th percentiles. If no fences are required set p = NULL.

df

the degrees of freedom for the Chi-square distribution, the number of variables in the multivariate distribution.

ifflip

by default fence labelling is placed to the left of the fences just above the x-axis. Setting ifflip = TRUE places the annotation to the right.

cex

the scale expansion factor for the fence labelling, by default cex = 0.6.

See Also

gx.md.plot, gx.md.gait, gx.md.gait.closed

Examples

# NOT RUN {
##  Synthesize test data
test <- mvrnorm(100, mu = c(40 ,30),Sigma = matrix(c(6, 3, 3, 2), 2, 2))

## Display annotated Chi-square plot
gx.md.gait(test)
gx.md.gait(test, ifadd = c(0.9, 0.98))

## Clean-up
rm(test)
# }

Run the code above in your browser using DataCamp Workspace