mvn.hyper.check: Argument checking for the default multivariate normal prior.

In msde: Bayesian Inference for Multivariate Stochastic Differential Equations

Description

Argument checking for the default multivariate normal prior.

Usage

mvn.hyper.check(hyper, param.names, data.names)

Arguments

hyper	The normal prior's hyperparameters: NULL, or a list with elements mu and Sigma, corresponding to a named mean vector and variance matrix (see Details).
param.names	Vector of parameter names (see Details).
data.names	Vector of data names (see Details).

Details

This function is not meant to be called directly by the user, but rather to parse the hyper-parameters of a default multivariate normal prior distribution to be passed to the C++ code in sde.prior() and sde.post(). This default prior is multivariate normal on the elements of (theta, x0) specified by each of names(mu), rownames(Sigma), and colnames(Sigma). The remaining components are given Lebesgue priors, or a full Lebesgue prior if hyper == NULL. If the names of mu and Sigma are inconsistent an error is thrown.

Value

A list with the following elements:

mean

The mean vector.

cholSd

The upper upper Cholesky factor of the variance matrix.

thetaId

The index of the corresponding variables in theta.

xId

The index of the corresponding variables in x0.

msde documentation built on Dec. 17, 2021, 9:07 a.m.