GHQ: Gaussian Hermite Quadrature with prunning

In GHQp: Gauss Hermite Quadrature with pruning.

Description

This function is used to obtain quadrature points to approximate an integral

Usage

GHQ(n, ndim, pruning = TRUE)

Arguments

n	number of quadrature points
ndim	number of integrals or dimension problem
pruning	a logical indicating whether you want pruning approach, by default is TRUE

Value

nodes	nodes
weights	weights
product	product weights

Author(s)

Freddy Hernandez Barajas

References

Hernandez, F., Usuga, O. and Giampaoli, V. (2014). Improving the Adaptive Gaussian Quadrature. Journal of Statistical Software, submitting.

Examples

# Comparing the number of points in a two-dimensional case
require(GHQp)
par(mfrow=c(2,2))
plot(GHQ(15,2,FALSE)$nodes,pch=20,xlab='',ylab='',
     main='Without pruning, n=15 and q=2')
plot(GHQ(15,2,TRUE)$nodes, pch=20,xlab='',ylab='',
     main='With pruning, n=15 and q=2')

# Comparing the number of points in a three-dimensional case
require(scatterplot3d)
datos <- GHQ(15,3,FALSE)$nodes
scatterplot3d(datos, type="p", highlight.3d=TRUE,
              angle=55, scale.y=0.7, pch=16,
              main='Without pruning, n=15 and q=3', 
              cex.symbols=0.4,xlab='',ylab='',zlab='')
datos <- GHQ(15,3,TRUE)$nodes
scatterplot3d(datos, type="p", highlight.3d=TRUE,
              angle=55, scale.y=0.7, pch=16,
              main='With pruning, n=15 and q=3', 
              cex.symbols=0.4,xlab='',ylab='',zlab='')

Example output

Loading required package: statmod
Loading required package: scatterplot3d

GHQp documentation built on May 2, 2019, 3:39 p.m.