init.gauss: Initialize Gauss-Hermite quadrature points and weights.

In Karel-Kroeze/MCAT: What the package does (short line)

Description

Uses fastGHQuad to create unidemsional quadrature points. These are subsequently expanded across the Q-dimensional grid, and shifted to account for the prior covariance matrix. Weights are computed as the product of weights for each dimension at each quadrature point. The mean vector is assumed to be zero, alternative prior means are not supported at this point.

Usage

init.gauss(ip = 10, prior)

Arguments

ip	Integer, number of quadrature points, per dimension. Note that the total number of quadrature points is ip^Q.
prior	Covariance matrix of the prior distribution.

Details

Details

Value

X Matrix of dimensions ip^Q by Q of quadrature points.

W Vector of length ip^Q of weights.

ip Integer number of quadrature points per dimension used after sanity check.

See Also

Package fastGHQuad, eval.gauss().

Karel-Kroeze/MCAT documentation built on May 8, 2019, 4:50 p.m.