rgeneric: rgeneric models

In inbo/INLA: Full Bayesian Analysis of Latent Gaussian Models using Integrated Nested Laplace Approximations

Description

A framework for defining latent models in R

Usage

inla.rgeneric.define(model = NULL, debug = FALSE, ...)
inla.rgeneric.iid.model(
        cmd = c("graph", "Q", "mu", "initial", "log.norm.const", "log.prior", "quit"),
        theta = NULL)
inla.rgeneric.ar1.model(
        cmd = c("graph", "Q", "mu", "initial", "log.norm.const", "log.prior", "quit"),
        theta = NULL)
inla.rgeneric.wrapper(
        cmd = c("graph", "Q", "mu", "initial", "log.norm.const", "log.prior", "quit"),
        model, theta = NULL)
inla.rgeneric.q(
        rmodel, 
        cmd = c("graph", "Q", "mu", "initial", "log.norm.const", "log.prior", "quit"),
        theta = NULL)

Arguments

model	The definition of the model; see inla.rgeneric.ar1.model
rmodel	The rgeneric model-object, the output of inla.rgeneric.define
debug	Logical. Turn on/off debugging
cmd	An allowed request
theta	Values of theta
...	Named list of variables that defines the environment of model
debug	Logical. Enable debug output

Value

This allows a latent model to be defined in R. See inla.rgeneric.ar1.model and inla.rgeneric.iid.model and the documentation for worked out examples of how to define latent models in this way. This will be somewhat slow and is intended for special cases and protyping. The function inla.rgeneric.wrapper is for internal use only.

Author(s)

Havard Rue [e-mail]

inbo/INLA documentation built on Dec. 6, 2019, 9:51 a.m.