inla.spde2.generic: Generic spde2 model creation.

In andrewzm/INLA: Functions which allow to perform full Bayesian analysis of latent Gaussian models using Integrated Nested Laplace Approximaxion

Description

Creates and inla.spde2 object describing the internal structure of an 'spde2' model.

Usage

inla.spde2.generic(M0, M1, M2, B0, B1, B2, theta.mu, theta.Q,
                   transform = c("logit", "log", "identity"),
                   theta.initial = theta.mu,
                   fixed = rep(FALSE, length(theta.mu)),
                   theta.fixed = theta.initial[fixed],
                   BLC = cbind(0, diag(nrow = length(theta.mu))),
                   ...)

## Map theta values to internal phi values
inla.spde2.theta2phi0(spde, theta)
inla.spde2.theta2phi1(spde, theta)
inla.spde2.theta2phi2(spde, theta)

Arguments

M0	The symmetric M0 matrix.
M1	The square M1 matrix.
M2	The symmetric M2 matrix.
B0	Basis definition matrix for phi0.
B1	Basis definition matrix for phi1.
B2	Basis definition matrix for phi2.
theta.mu	Prior expectation for the theta vector
theta.Q	Prior precision for the theta vector
transform	Transformation link for phi2. Valid settings are "logit", "log", and "identity"
theta.initial	Initial value for the theta vector. Default theta.mu
fixed	Logical vector. For every TRUE value, treat the corresponding theta value as known.
theta.fixed	Vector holding the values of fixed theta values. Default =theta.initial[fixed]
BLC	Basis definition matrix for linear combinations of theta.
...	Additional parameters, currently unused.
spde	An inla.sdpe2 object.
theta	parameter values to be mapped.

Value

For inla.spde2.generic, an inla.spde2 object.

For inla.spde2.theta2phi0/1/2, a vector of phi values.

Author(s)

Finn Lindgren [e-mail]

See Also

inla.spde2.models, inla.spde2.matern

andrewzm/INLA documentation built on May 10, 2019, 11:12 a.m.