dmvn_smw: dmvn_smw_mra
In jtipton25/BayesMRA: Bayesian Multi-Resolution Gaussian Process Approximations

dmvn_smw

R Documentation

dmvn_smw_mra

Description

dmvn_smw_mra

Usage

dmvn_smw(
  y,
  X,
  beta,
  tW,
  tWW,
  Q_alpha_tau2,
  sigma2,
  Rstruct = NULL,
  logd = TRUE
)

Arguments

`y`	is a `n` vector of Gaussian data.
`X`	is a `n \times p` matrix of fixed effects (like latitude, elevation, etc)
`beta`	is the regression parameter beta
`tW`	is the transpose of the sparse Wendland basis matrix of class spam
`tWW`	is the transpose of the sparse Wendland basis matrix multiplied by itself of class spam
`Q_alpha_tau2`	is the prior precision matrix for random effects alpha
`sigma2`	is the residual error
`Rstruct`	is the Cholesky prior precision matrix for random effects alpha
`logd`	is a logical value of whether to calculate the log density (`logd = TRUE`) or the density on the data scale (`logd = FALSE`)

Value

The (log) density of a normal distribution with mean \mathbf{X} \boldsymbol{\beta} and covariance matrix \sigma^2 \mathbf{I} + \mathbf{W} \mathbf{Q}_{\alpha_{\tau^2}} \mathbf{W}'

jtipton25/BayesMRA documentation built on Feb. 28, 2024, 1:27 p.m.