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

Table 1

List of true values of a’s, and 𝜇’s, 𝜎’s.

Sample name and size (in bracket)	Prior probability	Mean	Variance
Sample 1; Two components	𝑎1=0.4; 𝑎2=0.6	𝜇1=0.0; 𝜇2=2.0	𝜎^21=1.0; 𝜎^22=0.25
Sample 2; Two components	𝑎1=0.85; 𝑎2=0.15	𝜇1=0.0; 𝜇2=2.0	𝜎^21=1.0; 𝜎^22=0.25
Sample 3; Three components	𝑎1=0.33; 𝑎2=0.33, 𝑎3=0.34	𝜇1=0.0; 𝜇2=−1.0, 𝜇3=4.0	𝜎^21=1.0; 𝜎^22=0.25, 𝜎^23=4.0