Linear Discriminant Boundary

Share:

Description

Given two mean vectors and a covariance matrix (and optional prior probabilities), this function will return the slope and intercept of the boundary line between the two categories.

Usage

1
ldboundary (mean1, mean2, covariance, prior1 = .5, prior2 = .5, add = F, ...)

Arguments

mean1

A mean vector for category 1, must contain 2 elements.

mean2

A mean vector for category 2, must contain 2 elements.

covariance

A 2x2 covariance matrix for both distributions.

prior1

The prior probability of category 1.

prior2

The prior probability of category 2.

add

If TRUE, the boundary line is added top the plot.

...

Additional parameters are passed to the internal call of the line plotting function, in the event that add = T.

Value

The slope and intercept of the boundary line are returned.

Author(s)

Santiago Barreda <sbarreda@ucdavis.edu>

References

http://en.wikipedia.org/wiki/Linear_discriminant_analysis https://onlinecourses.science.psu.edu/stat557/book/export/html/35

Examples

1
2
3
4
5
6
7
8
9
## create two groups with the same covariance patterns
group1 = rmvtnorm (200, means= c(0,0), k=2, sigma = -.4)
group2 = rmvtnorm (200, means= c(3,3), k=2, sigma = -.4)
covariance = (var (group1) + var (group2)) / 2

## plot groups and boundary line between categories.
plot (group1, col = 2, pch = 16, ylim = c(-2,5), xlim = c(-2,5))
points (group2, col = 4, pch = 16)
ldboundary (c(0,0), c(3,3), covariance, add = TRUE)

Want to suggest features or report bugs for rdrr.io? Use the GitHub issue tracker.