rtcmvnorm: Random Generation from a Truncated Conditional Normal...

rtcmvnormR Documentation

Random Generation from a Truncated Conditional Normal Distribution

Description

Samples from the Truncated Conditional Normal Distribution

Usage

rtcmvnorm(n = 1, mean = 0, V = 1, x=0, keep=1, lower = -Inf, upper = Inf)

Arguments

n

integer: number of samples to be drawn

mean

vector of means

V

covariance matrix

x

vector of observations to condition on

keep

element of x to be sampled

lower

left truncation point

upper

right truncation point

Value

vector

Author(s)

Jarrod Hadfield j.hadfield@ed.ac.uk

Examples

par(mfrow=c(2,1))
V1<-cbind(c(1,0.5), c(0.5,1))
x1<-rtcmvnorm(10000, c(0,0), V=V1, c(0,2), keep=1, lower=-1, upper=1)
x2<-rtnorm(10000, 0, 1, lower=-1, upper=1)
plot(density(x1), main="Correlated conditioning observation")
lines(density(x2), col="red")
# denisties of conditional (black) and unconditional (red) distribution
# when the two variables are correlated (r=0.5) 

V2<-diag(2)
x3<-rtcmvnorm(10000, c(0,0), V=V2, c(0,2), keep=1, lower=-1, upper=1)
x4<-rtnorm(10000, 0, 1, lower=-1, upper=1)
plot(density(x3), main="Uncorrelated conditioning observation")
lines(density(x4), col="red")
# denisties of conditional (black) and unconditional (red) distribution
# when the two variables are uncorrelated (r=0) 


MCMCglmm documentation built on July 9, 2023, 5:24 p.m.

Related to rtcmvnorm in MCMCglmm...