phonTools (version 0.1-6)

sdellipse: Standard deviation Ellipse

Description

Draw standard deviation ellipses around a group of observations.

Usage

sdellipse (points, stdev = 1.96, density = .1, output = FALSE, add = TRUE, show = TRUE, means = NULL, ...)

Arguments

points

A matrix with two columns in which each row is a different observation from a bivariate normal distribution. Optionally, a 2 by 2 covariance matrix may be specified directly in conjuction with the means parameter.

stdev

The number of standard deviations to be enclosed by the ellipse.

density

The spacing between sampling points along the ellipse. A higher number results in a coarser sampling.

output

If TRUE, the points used to draw the ellipse are returned.

add

If TRUE, the ellipse is added to an existing plot. If FALSE, a new plot is created.

show

If FALSE, no ellipse is drawn.

means

A vector of 2 means, one for each dimension. If these are specified, points is assumed to be a covariance matrix rahter than a sequence of observations.

...

Additional arguments are passed to the internal call of lines() or plot() as appropriate.

Value

• If output = TRUE, a matrix is returned where the first column indicate x-axis ellipse coordinates and the second column indicates y-axis ellipse coordinates.

Examples

## Examples of draws from different bivariate normal distributions
## and standard deviation ellipses drawn to fit them.
par (mfrow = c(2,2))
draws = rmvtnorm (n = 10000, k = 2, sigma = .3)
plot (draws)
sdellipse (draws, stdev = 3, lwd = 3, col = 2)

draws = rmvtnorm (n = 10000, k = 2, sigma = -.3)
plot (draws)
sdellipse (draws, stdev = 3, lwd = 3, col = 2)

draws = rmvtnorm (n = 10000, k = 2, sigma = -.7)
plot (draws)
sdellipse (draws, stdev = 3, lwd = 3, col = 2)

draws = rmvtnorm (n = 10000, k = 2, sigma = .7)
plot (draws)
sdellipse (draws, stdev = 3, lwd = 3, col = 2)

## alternatively, a covariance matrix may be specified directly. 
par (mfrow = c(1,1))
sdellipse (matrix(c(1,.5,.5,1),2,2), means = c(0,0), add = FALSE)
sdellipse (matrix(c(1,-.5,-.5,1),2,2), means = c(0,0), add = TRUE)