compute_NEC: Compute Negative Entropy Criterion (NEC)

In LICORS: Light Cone Reconstruction of States - Predictive State Estimation From Spatio-Temporal Data

Description

Computes the negative entropy criterion (NEC) to assess the number of clusters in a mixture model. See References for details.

Usage

compute_NEC(weight.matrix, loglik.1 = NULL, loglik.k = NULL)

Arguments

weight.matrix	N \times K weight matrix
loglik.1	baseline log-likelihood for K=1 cluster model
loglik.k	log-likelihood for K cluster model

References

Christophe Biernacki, Gilles Celeux, and G\'erand Govaert(1999). “An improvement of the NEC criterion for assessing the number of clusters in a mixture model”. Non-Linear Anal. 20, 3, 267-272.

See Also

compute_mixture_penalty

Examples

WW <- matrix(c(rexp(10, 1/10), runif(10)), ncol = 5, byrow = FALSE)
WW <- normalize(WW)
compute_NEC(WW, -2, -1)

LICORS documentation built on May 1, 2019, 10:13 p.m.