simulate.moment: Method to compute a multivariate moment using Monte Carlo...

In symmoments: Symbolic Central and Noncentral Moments of the Multivariate Normal Distribution

Description

Computes a multivariate normal moment by Monte Carlo integration

Usage

## S3 method for class 'moment'
simulate(object,nsim,seed,Mean,Sigma,...)

Arguments

object	object of class 'moment' representing E[X1**k1,...,Xn**kn]
nsim	the number of samples to generate in computing the integral
seed	integer for random number generator (set.seed)
Mean	the mean of (X1,...,Xn)
Sigma	covariance of (X1**k1,...,Xn**kn), dimension nXn, expressed as a vector by row
...	Included only for consistency with generic function

Value

Approximate value of the moment

Note

Non-central moments can be approximated by specifying Mean. For central moments, set Mean to a vector of 0s.

The mvtnorm package must be loaded for the function rmvnorm.

Author(s)

Kem Phillips <[e-mail]>

References

Rizzo ML (2008). Statistical Computing with R. Chapman & Hall/CRC

See Also

callmultmoments and the methods toLatex and evaluate from symmoments

Examples

# Using 10000 samples, estimate the central moment for the moment c(2,4) at the covariance matrix
#  2 1
#  1 4

# and mean (0,0)
library(mvtnorm)
simulate(callmultmoments(c(2,4)),10000,NULL,c(0,0),c(2,1,1,4))

Example output

Loading required package: mvtnorm
Loading required package: cubature
Loading required package: combinat

Attaching package: ‘combinat’

The following object is masked from ‘package:utils’:

    combn

Loading required package: multipol
Loading required package: abind

Attaching package: ‘multipol’

The following object is masked from ‘package:stats’:

    deriv

The following object is masked from ‘package:base’:

    single

[1] 156.8237

symmoments documentation built on March 26, 2020, 6:28 p.m.