devpart: Deviance partitioning

Description

Runs a deviance partitioning procedure on a set of four bayescomm objects.

Usage

devpart(null, environment, community, full)

Arguments

null

a bayescomm object containing a 'null' model

environment

a bayescomm object containing an 'environment' model

community

a bayescomm object containing a 'community' model

full

a bayescomm object containing a 'full' model

Value

devpart: matrix containing the proportion of the null deviance explained by each model for each species
null: a matrix containing the mean and 95% credible intervals for the deviance for each species in the null model
environment: a matrix containing the mean and 95% credible intervals for the deviance for each species in the evironment model
community: a matrix containing the mean and 95% credible intervals for the deviance for each species in the community model
full: a matrix containing the mean and 95% credible intervals for the deviance for each species in the full model

Details

The deviance partitioning procedure determines the proportion of the null deviance explained by each of the other three model types. The four model types are those created by BC.

Examples

Run this code

# create fake data
n <- 100
nsp <- 4
k <- 3

X <- matrix(c(rep(1, n), rnorm(n * k)), n)  # covariate matrix
W <- matrix(rnorm(nsp * nsp), nsp)
W <- W %*% t(W) / 2  # true covariance matrix
B <- matrix(rnorm(nsp * (k + 1), 0, 3), nsp)  # true covariates
mu <- apply(B, 1, function(b, x) x %*% b, X)  # true mean
e <- matrix(rnorm(n * nsp), n) %*% chol(W)  # true e
z <- mu + e  # true z
Y <- ifelse(z > 0, 1, 0)  # true presence/absence

# run BC (after removing intercept column from design matrix)
null <- BC(Y, X[, -1], model = "null", its = 100)
comm <- BC(Y, X[, -1], model = "community",its = 100)
envi <- BC(Y, X[, -1], model = "environment", its = 100)
full <- BC(Y, X[, -1], model = "full", its = 100)

devpart(null, envi, comm, full)