DIC: Deviance Information Criterion
In BayesComm: Bayesian Community Ecology Analysis

Description Usage Arguments References See Also Examples

Calculates Deviance Information Criteria for bayescomm objects.

DIC(BC)

`BC`	a `bayescomm` object

Spiegelhalter, D.J.. Best, N.G., Carlin, B.P., van der Linde, A. (2002). Bayesian measures of model complexity and fit. Journal of the Royal Statistical Society, Series B, 64 (4): 583-639.

BC

1 2	m1 <- example(BC)[[1]] DIC(m1)

BC> # create fake data
BC> n <- 100

BC> nsp <- 4

BC> k <- 3

BC> X <- matrix(c(rep(1, n), rnorm(n * k)), n)  # covariate matrix

BC> W <- matrix(rnorm(nsp * nsp), nsp)

BC> W <- W %*% t(W) / 2  # true covariance matrix

BC> B <- matrix(rnorm(nsp * (k + 1), 0, 3), nsp)  # true covariates

BC> mu <- apply(B, 1, function(b, x) x %*% b, X)  # true mean

BC> e <- matrix(rnorm(n * nsp), n) %*% chol(W)  # true e

BC> z <- mu + e  # true z

BC> Y <- ifelse(z > 0, 1, 0)  # true presence/absence

BC> # run BC (after removing intercept column from design matrix)
BC> m1 <- BC(Y, X[, -1], model = "full", its = 100)
[1] 115.7853