LogConcDEAD (version 1.2-0)

lcd.marg: Finds marginals of multivariate logconcave MLEs by integrating

Description

This function integrates the MLE of multivariate data to produce axis-aligned marginals for use in plot.LogConcDEAD.

Usage

lcd.marg (out, marg=1, gridlen=100)

Arguments

out

Output from lcd.mle (of class LogConcDEAD)

marg

Which margin is required?

gridlen

Size of grid

Value

• xo
  x-values
• marg
  Integrated MLE

Details

Given a multivariate log-concave MLE in the form of an object of class LogConcDEAD and a margin marg, this will integrate the MLE over the relevant subspaces along a line with gridlen points.

The R code for computing the function J was provided by Lutz D"umbgen.

Examples

#simple normal data, a couple of plots

set.seed(101)
x <- matrix(rnorm(200),ncol=2)
out <- lcd.mle(x)
m1 <- lcd.marg(out,marg=1)
m2 <- lcd.marg(out,marg=2)
par(mfrow=c(1,2))
plot(out,marg=1,g.marg=m1)
plot(out,marg=2,g.marg=m2)