Command Reference : Command Reference
  
 
rmvnorm
Generate multivariate random normal values.
Fill group of series with multivariate normals random draws.
Syntax
rmvnorm(x, S[, prefix]))
rmvnormc(x, S[, prefix]))
rmvnormi(x, S[, prefix]))
mvnormci(x, S[, prefix]))
where x is an object as described below matrix, and sym S describes the covariance matrix as described below.
If x doesn't exist, a group object is first created and populated with series named “<prefix>1”, “<prefix>2”, and so on. If omitted, the default prefix is "SER".
If x is a group, the command fills each observation of the contained series with a draw from the distribution.
If x is a vector, the command fills the vector with a single draw from the distribution.
If x is a matrix, the command fills each row of the matrix with a draw from the distribution.
There are four distinct commands in this family which correspond to different interpretations for S. These forms are distinguished by the characters in the command name that follow the initial string “rmvnorm”:
 
“”
Supply , the covariance matrix of distribution.
“c”
Supply the Cholesky decomposition of . This form is more efficient when performing multiple draws from the same distribution (compute the Cholesky once, but sample many times).
“i “
Supply . This form is more efficient than explicitly inverting to supply .
“ic”
Supply the Cholesky decomposition of . This form combines the efficiencies of the above forms.
Examples
sym a = @inner(@mnrnd(100, 10))
matrix(1000, 10) x
rmvnorm(x, a)
rmvnormc(x, @cholesky(a))
rmvnormi(x, @inverse(a))
rmvnormic(x, @cholesky(@inverse(a))
Cross-references
For random generator functions, see “Statistical Distributions” and in particular, @mrnd, @mnrnd, and @rmvnorm.
For related commands, see nrnd, rnd, rndint, and rmvnorm. See also rndseed.