................................................... MIXPDTA CLUSPAC - MIXTURE-MODEL CLUSTERING OF CASES VARYING COVARIANCE MATRICES AUTOMATIC SETTING OF INITIAL PARAMETER ESTIMATES Copyright (C) 1992 Stanley Louis Sclove Developed and programmed by: Prof. Stanley L. Sclove, Ph.D. Phone (312) 996-2676 Information & Decision Sciences Dept. (MC 294) University of Illinois at Chicago 601 S Morgan St. Chicago, IL 60607-7124 312-996-2676 slsclove@uic.edu www.uic.edu/~slsclove ................................................... MIXPDTA VERSION 3.01 2003: Mar 19 ..................................................... PROBLEM TITLE IS IRIS DATA (Sources: Primary - Anderson; Secondary - Fisher) NUMBER OF VARIABLES . . . . . . . . . . . . 4 NUMBER OF OBSERVATIONS (SAMPLE SIZE), N . . 150 Level of verbosity is 1 . IVERB = 1: terse output upon convergence, values of BIC only IVERB = 2: medium output above, plus means IVERB = 3: verbose output above, plus covariance matrices IVERB = 4: complete output above, for every iteration If KLO=KHI, LABELS will be written. MINIMUM OF EACH VARIABLE: 4.3 2.0 1.0 0.1 MAXIMUM OF EACH VARIABLE: 7.9 4.4 6.9 2.5 FIRST FOUR DATA POINTS ----- ---- ---- ------ 5.1 3.5 1.4 0.2 1 001 4.9 3.0 1.4 0.2 1 002 4.7 3.2 1.3 0.2 1 003 4.6 3.1 1.5 0.2 1 004 NUMBER OF CLUSTERS TO REPORT. . . . . . . . 1 TO 9 STATISTICS FOR WHOLE (UNCLUSTERED) SAMPLE: MEAN VECTOR: 5.84 3.06 3.76 1.20 COVARIANCE MATRIX: 0.681 -0.042 1.266 0.513 -0.042 0.189 -0.327 -0.121 1.266 -0.327 3.096 1.287 0.513 -0.121 1.287 0.577 NUMBER OF PARAMETERS = 14 MINUS 2 LOG LIKELIHOOD = 759.8 AIC = 787.8 BIC = 830.0 ..................................................... K = 2 CLUSTERS ..................................................... K = 3 CLUSTERS ..................................................... K = 4 CLUSTERS ..................................................... K = 5 CLUSTERS ..................................................... K = 6 CLUSTERS ..................................................... K = 7 CLUSTERS ..................................................... K = 8 CLUSTERS ..................................................... K = 9 CLUSTERS ..................................................... IRIS DATA (Sources: Primary - Anderson; Secondary - Fisher) MODEL SELECTION CRITERIA MODEL-SELECTION CRITERIA NUMBER MINUS 2 NUMBER (PPH=POST.PROB.OF MODEL): OF TIMES OF AKAIKE SCHWARZ CLUSTERS, LOG OF PARAM- ------------ ------------ K LIKELIHOOD ETERS VALUE PPH VALUE PPH --------- ---------- ----- ------- --- ------- --- 1 759.8 14 787.8 1.00 830.0 1.00 2 680.1 29 738.1 0.00 825.4 0.00 3 1052.7 44 1140.7 0.00 1273.2 0.00 4 1166.3 59 1284.3 0.00 1461.9 0.00 5 1738.3 74 1886.3 0.00 2109.1 0.00 6 2290.8 89 2468.8 0.00 2736.7 0.00 7 2421.4 104 2629.4 0.00 2942.5 0.00 8 2422.9 119 2660.9 0.00 3019.1 0.00 9 3782.2 134 4050.2 0.00 4453.6 0.00 ..................................................... IRIS DATA (Sources: Primary - Anderson; Secondary - Fisher) CONDITION CODE FOR THIS RUN: PROGRAM ENDED NORMALLY. ..................................................... MIXPDTA CLUSPAC - MIXTURE-MODEL CLUSTERING OF CASES TIME BEGAN: TIME FINISHED: