Gaussian Mixture Models Based on Principal Components and Applications

Table 3

Simulation case: parameter estimates of the bivariate GMM (second technique), taking cluster means as initial values of model means.

Sample size	Model parameters
	First component					Second component
	500	0.83586	0.08308	0.06959	2.99632	0.93938	0.16414	−0.4231	−0.35438	0.59144	1.61035
Log-likelihood = –1673.792, BIC = 3389.031; Number of iterations: 734