README.md

In jacsor/MPG: Mixtures of Perturbed Gaussians

Variability decomposition across related mixture distributions

This package fits MPG, an algorithm to compare multiple related mixture distributions.

Install

The package can be installed on Linux and Mac using devtools:

install.packages('devtools')
library('devtools')
devtools::install_github('MPG', 'jacsor')

Use

There are three functions in this package, and their descriptions are provided in the help files

ans = fit(Y, C)
plotDiff(ans)
cal = calibrate(ans)

Example

n = c(250, 250)
p = 4
Y1 = rbind( matrix( rnorm( n[1]*p), ncol = p), matrix( rnorm(n[2]*p) + 3, ncol = p))
Y2 = rbind( matrix( rnorm( n[1]*p), ncol = p), matrix( rnorm(n[2]*p) + 4, ncol = p))
Y = rbind(Y1, Y2)
C = c( rep(1,sum(n)), rep(2,sum(n)))
ans = mpg(Y, C)  
plotDiff(ans, type = "weight")
plotDiff(ans, type = "shift")
cal = calibrate(ans)
par(mfrow=c(1,2))
plot(Y, col = C)
plot(cal$Y_cal, col = C)

Reference

To do.

jacsor/MPG documentation built on May 18, 2019, 9:05 a.m.