mnormt (version 1.2-0)

dmnorm: Multivariate normal distribution

Description

The probability density function, the distribution function and random number generation for the multivariate normal (Gaussian) probability distribution

Usage

dmnorm(x, mean = rep(0, d), varcov, log = FALSE) 
pmnorm(x, mean = rep(0, length(x)), varcov, ...) 
rmnorm(n = 1, mean = rep(0, d), varcov) 
sadmvn(lower, upper, mean, varcov, maxpts = 2000 * d, abseps = 1e-06, releps = 0)

Arguments

x

for dmnorm, this is either a vector of length d or a matrix with d columns, where d=ncol(varcov), giving the coordinates of the point(s) where the density must be evalua

mean

a numeric vector representing the expected value of the distribution; it must be of length d, as defined above

varcov

a positive definite matrix representing the variance-covariance matrix of the distribution; a vector of length 1 is also allowed (in this case, d=1 is set)

log

a logical value; if TRUE, the logarithm of the density is computed

...

parameters passed to sadmvn, among maxpts, absrel, releps

n

the number of random numbers to be generated

lower

a numeric vector of lower integration limits of the density function; must be of maximal length 100; +Inf and -Inf entries are allowed

upper

a numeric vector of upper integration limits of the density function; must be of maximal length 100; +Inf and -Inf entries are allowed

maxpts

the maximum number of function evaluations (default value: 2000*d)

abseps

absolute error tolerance (default value: 1e-6)

releps

relative error tolerance (default value: 0)

Value

• dmnorm returns a vector of density values (possibly log-transformed); pmnorm and sadmvn return a single probability with attributes giving details on the achieved accuracy; rmnorm returns a matrix of n rows of random vectors

Details

Function sadmvn is an interface to a Fortran-77 routine with the same name written by Alan Genz, and available from his web page; this makes uses of some auxiliary functions whose authors are documented in the Fortran code. The routine uses an adaptive integration method.

References

Genz, A. (1992). Numerical Computation of Multivariate Normal Probabilities. J. Computational and Graphical Statist., 1, 141-149. Genz, A. (1993). Comparison of methods for the computation of multivariate normal probabilities. Computing Science and Statistics, 25, 400-405. Genz, A.: Fortran code available at http://www.math.wsu.edu/math/faculty/genz/software/mvn.f

See Also

dnorm, dmt

Examples

x <- seq(-2,4,length=21)
y <- 2*x+10
z <- x+cos(y) 
mu <- c(1,12,2)
Sigma <- matrix(c(1,2,0,2,5,0.5,0,0.5,3), 3, 3)
f <- dmnorm(cbind(x,y,z), mu, Sigma)
p1 <- pmnorm(c(2,11,3), mu, Sigma)
p2 <- pmnorm(c(2,11,3), mu, Sigma, maxpts=10000, abseps=1e-10)
x <- rmnorm(10, mu, Sigma)
p <- sadmvn(lower=c(2,11,3), upper=rep(Inf,3), mu, Sigma) # upper tail
#
p1 <- pnorm(0, 1, 2)
p2 <- pmnorm(0, 1, 2^2)