Example 2.8 Figure 2.17a shows N=100 points in the two-dimensional space, from a multimodal distribution. The samples were generated using two Gaussian random generators N(mu_(1),Sigma_(1)),N(mu_(2),Sigma_(2)), with mu_(1)=[[1.0],[1.0]],mu_(2)=[[2.0],[2.0]] and covariance matrices Sigma_(1)=Sigma_(2)=[[0.1,0.0],[0.0,0.1]] respectively. Each time a sample x_(k),k=1,2,dots,N, is to be generated a coin is tossed. The corresponding probabilities for heads or tails are P(H)-=P=0.8,P(T)=1-P=0.2, respec