Improved Expectation Maximization Algorithm for Gaussian Mixed Model Using the Kernel Method

Table 3

Simulation results for the case 𝜇₁=0, 𝜇₂=1.0, 𝜎²₂=0.707, and 𝜎²₁=0.447.

Prior prob.		Bias, 𝐷𝑖
		𝑎1		𝑎2		𝜇1		𝜇2		𝜎^21		𝜎^22
𝑎1	𝑎2	𝐷	𝑆𝐷	𝐷	𝑆𝐷	𝐷	𝑆𝐷	𝐷	𝑆𝐷	𝐷	𝑆𝐷	𝐷	𝑆𝐷
0.1	0.9	−0.112	0.148	0.112	0.148	−0.130	0.203	−0.067	0.139	−0.030	0.103	0.034	0.068
0.2	0.8	−0.014	0.073	0.014	0.073	−0.006	0.087	0.006	0.091	0.022	0.056	0.005	0.057
0.3	0.7	0.020	0.087	−0.020	0.087	0.031	0.069	0.023	0.104	0.044	0.048	−0.006	0.075
0.4	0.6	0.067	0.075	−0.067	0.075	−0.112	0.444	0.145	0.232	−0.002	0.085	−0.021	0.099