gev.update.M: Sample a new model from the current model for any linear...

Description Usage Arguments Value Author(s)

View source: R/gev.R

Description

This uses a conditional Bayes factor (CBF) to update a model in a linear system given a current model and other information in a spatial GEV model. Note that it is agnostic to which part of the framework (location, precision, scale) you are updating.

Usage

1
gev.update.M(Y, X, M, alpha, lambda, D, beta.0, Omega.0)

Arguments

Y

The current dependent variable, calculated relative to the linear plus random effect terms of the given component.

X

The matrix of covariates

M

The current model. A subset of (1, ..., p) where p is the number of columns in X

alpha

The precision term of the Gaussian process for this component of the model

lambda

The length term of the Gaussian process for this component of the model

D

The distance matrix used in the Gaussian process

beta.0

The prior mean on the linear regression terms

Omega.0

The prior covariance on the linear regression terms

Value

This returns an updated model, which is a vector that is a subset of (1, ..., p).

Author(s)

Alex Lenkoski <[email protected]>


spatial.gev.bma documentation built on May 29, 2017, 9:10 a.m.