omxMnor {OpenMx}R Documentation

Multivariate Normal Integration


Given a covariance matrix, a means vector, and vectors of lower and upper bounds, returns the multivariate normal integral across the space between bounds.


omxMnor(cov, means, lbounds, ubounds)


cov the covariance matrix describing the multivariate normal distribution.
means a row vector containing means of the variables of the underlying distribution.
lbounds a row vector containing the lower bounds of the integration in each variable.
ubounds a row vector containing the upper bounds of the integration in each variable.


The order of columns in the means, lbounds, and ubounds vector are assumed to be the same as that of the covariance matrix. That is, means[i] is considered to be the mean of the variable whose variance is in cov[i,i]. That variable will be integrated from lbounds[i] to ubounds[i] as part of the integration.

The value of ubound[i] or lbound[i] may be set to Inf or -Inf if a boundary at positive or negative infinity is desired.

For all i, ubound[i] must be strictly greater than lbound[i].



cov <- cov(myFAData[,1:3])
means <- mean(myFAData[,1:3])
lbounds <- c(-Inf, 0,   1)    # Integrate from -Infinity to 0 on first variable
ubounds <- c(0,    Inf, 2.5)  # From 0 to +Infinity on second, and from 1 to 2.5 on third
omxMnor(cov, means, lbounds, ubounds)

[Package OpenMx version 0.2.9-1147 Index]