pcbic function

BIC for a particular pattern

Find the BIC and MLE from a set of observed eigenvalues for a specific pattern.

pcbic(eigenvals, n, pattern)

Arguments

• eigenvals: The $Q$-vector of eigenvalues of the covariance matrix, in order from largest to smallest.
• n: The degrees of freedom in the covariance matrix.
• pattern: The pattern of equalities of the eigenvalues, given by the $K$-vector ($Q_1$, ... , $Q_K$) as in (13.8).

Returns

A list with the following components:

• lambdaHat: A $Q$-vector containing the MLE's for the eigenvalues.
• Deviance: The deviance of the model, as in (13.13).
• Dimension: The dimension of the model, as in (13.12).
• BIC: The value of the BIC for the model, as in (13.14).

Examples

# Build cars1
require("mclust")
mcars <- Mclust(cars)
cars1 <- cars[mcars$classification == 1, ]
xcars <- scale(cars1)
eg <- eigen(var(xcars))
pcbic(eg$values, 95, c(1, 1, 3, 3, 2, 1))

See Also

pcbic.stepwise, pcbic.unite, and pcbic.subpatterns.