mdir (version 0.9.0)

calcAllocProb: Calculate allocation probabilities

Description

Calculate the empirical allocation probability for each class based on the sampled allocation probabilities. Only makes sense in semi-supervised views.

Usage

calcAllocProb(mcmc_samples, view, burn = 0, method = "mean")

Value

An N x K matrix of class probabilities.

Arguments

mcmc_samples

Output from ``callMDI``.

view

The view for which to calculate the allocation probabilities.

burn

The number of samples to discard.

method

The point estimate to use. ``method = 'mean'`` or ``method = 'median'``. ``'median'`` is the default.

Examples

N <- 100
X <- matrix(c(rnorm(N, 0, 1), rnorm(N, 3, 1)), ncol = 2, byrow = TRUE)
Y <- matrix(c(rnorm(N, 0, 1), rnorm(N, 3, 1)), ncol = 2, byrow = TRUE)

truth <- c(rep(1, N / 2), rep(2, N / 2))
data_modelled <- list(X, Y)

V <- length(data_modelled)

# These values are much too low for real applications
R <- 50
thin <- 5
burn <- 10

alpha <- rep(1, V)
K_max <- 10
K <- rep(K_max, V)
labels <- matrix(1, nrow = N, ncol = V)
fixed <- matrix(0, nrow = N, ncol = V)

# A random quarter of labels are known in view 1
fixed[, 1] <- sample(c(0, 1), N, replace = TRUE, prob = c(3, 1))
labels[, 1] <- generateInitialSemiSupervisedLabels(truth, fixed = fixed[, 1])
types <- rep("G", V)

mcmc_out <- callMDI(data_modelled, R, thin, types, K, labels, fixed, alpha)
calcAllocProb(mcmc_out, 1)