1................................................. PROGRAM MIX1DT CLUSPAC MIXTURE MODEL CLUSTERING FOR UNIVARIATE DATA (DATA ON THE LINE) WITH UNEQUAL CLASS VARIANCES DEVELOPED AND PROGRAMMED BY DR. STANLEY L. SCLOVE VERSION 1.6 15-MAY-92 CMS DSN = MIX1DT CLUSPAC COPYRIGHT (C) 1991, 1992 STANLEY L. SCLOVE SIMULATED BETA MIXTURE: 299 (1,8), 400 (3,4), 300 (7,3) N = 999 DATA: MINIMUM OF SAMPLE: 0.000592 MAXIMUM OF SAMPLE: 0.985508 MEAN = 0.4061 M.L. ESTIMATE OF VARIANCE = 0.07080 SSDEVS = 70.7244 MINUS 2 LOG LIKELIHOOD = 189.7235 STDDEV = 0.2661 AIC = 193.7235 KASHYAP CRITERION = 210.7877 1 K = 3 CLUSTERS INITIAL PRIOR PROBS, MEANS AND VARIANCES: 1 0.30 0.17 0.01 2 0.40 0.43 0.04 3 0.30 0.77 0.02 WGSS = 22.8663 MINUS 2 LOG LIKELIHOOD = 79.5631 WGMS = 0.0230 STD.ERROR=SQRT(WGMS) = 0.1515 ITERATION 1 BOUNDARIES: 0.260 0.615 MEANS: 0.138 0.426 0.720 WGSS = 21.1860 MINUS 2 LOG LIKELIHOOD = -17.2260 WGMS = 0.0213 STD.ERROR=SQRT(WGMS) = 0.1458 ITERATION 2 BOUNDARIES: 0.257 0.615 MEANS: 0.126 0.432 0.718 WGSS = 19.6612 MINUS 2 LOG LIKELIHOOD = -32.9026 WGMS = 0.0197 STD.ERROR=SQRT(WGMS) = 0.1405 ITERATION 3 BOUNDARIES: 0.255 0.616 MEANS: 0.117 0.434 0.721 WGSS = 18.3323 MINUS 2 LOG LIKELIHOOD = -47.4950 WGMS = 0.0184 STD.ERROR=SQRT(WGMS) = 0.1357 ITERATION 4 BOUNDARIES: 0.251 0.617 MEANS: 0.109 0.434 0.724 WGSS = 17.2286 MINUS 2 LOG LIKELIHOOD = -61.2207 WGMS = 0.0173 STD.ERROR=SQRT(WGMS) = 0.1315 ITERATION 5 BOUNDARIES: 0.247 0.616 MEANS: 0.103 0.433 0.728 WGSS = 16.3679 MINUS 2 LOG LIKELIHOOD = -73.1920 WGMS = 0.0164 STD.ERROR=SQRT(WGMS) = 0.1282 ITERATION 6 BOUNDARIES: 0.242 0.615 MEANS: 0.098 0.430 0.732 WGSS = 15.7475 MINUS 2 LOG LIKELIHOOD = -82.8066 WGMS = 0.0158 STD.ERROR=SQRT(WGMS) = 0.1257 ITERATION 7 BOUNDARIES: 0.237 0.613 MEANS: 0.093 0.426 0.735 WGSS = 15.3439 MINUS 2 LOG LIKELIHOOD = -89.9856 WGMS = 0.0154 STD.ERROR=SQRT(WGMS) = 0.1241 ITERATION 8 BOUNDARIES: 0.232 0.611 MEANS: 0.089 0.423 0.737 WGSS = 15.1207 MINUS 2 LOG LIKELIHOOD = -95.1194 WGMS = 0.0152 STD.ERROR=SQRT(WGMS) = 0.1232 ITERATION 9 BOUNDARIES: 0.228 0.610 MEANS: 0.086 0.419 0.739 WGSS = 15.0370 MINUS 2 LOG LIKELIHOOD = -98.7854 WGMS = 0.0151 STD.ERROR=SQRT(WGMS) = 0.1229 ITERATION 10 BOUNDARIES: 0.223 0.608 MEANS: 0.084 0.415 0.739 WGSS = 15.0545 MINUS 2 LOG LIKELIHOOD = -101.5067 WGMS = 0.0151 STD.ERROR=SQRT(WGMS) = 0.1229 ITERATION 11 BOUNDARIES: 0.219 0.606 MEANS: 0.082 0.412 0.740 WGSS = 15.1409 MINUS 2 LOG LIKELIHOOD = -103.6472 WGMS = 0.0152 STD.ERROR=SQRT(WGMS) = 0.1233 ITERATION 12 BOUNDARIES: 0.215 0.605 MEANS: 0.080 0.409 0.740 WGSS = 15.2715 MINUS 2 LOG LIKELIHOOD = -105.4301 WGMS = 0.0153 STD.ERROR=SQRT(WGMS) = 0.1238 ITERATION 13 BOUNDARIES: 0.211 0.603 MEANS: 0.078 0.406 0.739 WGSS = 15.4281 MINUS 2 LOG LIKELIHOOD = -106.9743 WGMS = 0.0155 STD.ERROR=SQRT(WGMS) = 0.1245 ITERATION 14 BOUNDARIES: 0.207 0.602 MEANS: 0.076 0.403 0.739 WGSS = 15.5977 MINUS 2 LOG LIKELIHOOD = -108.3490 WGMS = 0.0157 STD.ERROR=SQRT(WGMS) = 0.1251 ITERATION 15 BOUNDARIES: 0.204 0.601 MEANS: 0.075 0.400 0.739 WGSS = 15.7718 MINUS 2 LOG LIKELIHOOD = -109.5912 WGMS = 0.0158 STD.ERROR=SQRT(WGMS) = 0.1258 ITERATION 16 BOUNDARIES: 0.200 0.600 MEANS: 0.073 0.398 0.738 WGSS = 15.9444 MINUS 2 LOG LIKELIHOOD = -110.7255 WGMS = 0.0160 STD.ERROR=SQRT(WGMS) = 0.1265 ITERATION 17 BOUNDARIES: 0.197 0.598 MEANS: 0.072 0.396 0.738 WGSS = 16.1121 MINUS 2 LOG LIKELIHOOD = -111.7680 WGMS = 0.0162 STD.ERROR=SQRT(WGMS) = 0.1272 ITERATION 18 BOUNDARIES: 0.194 0.597 MEANS: 0.071 0.393 0.737 WGSS = 16.2725 MINUS 2 LOG LIKELIHOOD = -112.7323 WGMS = 0.0163 STD.ERROR=SQRT(WGMS) = 0.1278 ITERATION 19 BOUNDARIES: 0.190 0.596 MEANS: 0.070 0.391 0.737 WGSS = 16.4245 MINUS 2 LOG LIKELIHOOD = -113.6287 WGMS = 0.0165 STD.ERROR=SQRT(WGMS) = 0.1284 ITERATION 20 BOUNDARIES: 0.187 0.595 MEANS: 0.068 0.389 0.736 WGSS = 16.5675 MINUS 2 LOG LIKELIHOOD = -114.4673 WGMS = 0.0166 STD.ERROR=SQRT(WGMS) = 0.1290 ITERATION 21 BOUNDARIES: 0.184 0.594 MEANS: 0.067 0.387 0.736 WGSS = 16.7016 MINUS 2 LOG LIKELIHOOD = -115.2557 WGMS = 0.0168 STD.ERROR=SQRT(WGMS) = 0.1295 ITERATION 22 BOUNDARIES: 0.181 0.593 MEANS: 0.066 0.385 0.735 WGSS = 16.8268 MINUS 2 LOG LIKELIHOOD = -116.0028 WGMS = 0.0169 STD.ERROR=SQRT(WGMS) = 0.1300 ITERATION 23 BOUNDARIES: 0.179 0.592 MEANS: 0.065 0.384 0.734 WGSS = 16.9438 MINUS 2 LOG LIKELIHOOD = -116.7139 WGMS = 0.0170 STD.ERROR=SQRT(WGMS) = 0.1304 ITERATION 24 BOUNDARIES: 0.176 0.591 MEANS: 0.064 0.382 0.734 WGSS = 17.0531 MINUS 2 LOG LIKELIHOOD = -117.3958 WGMS = 0.0171 STD.ERROR=SQRT(WGMS) = 0.1308 ITERATION 25 BOUNDARIES: 0.173 0.590 MEANS: 0.063 0.380 0.733 WGSS = 17.1551 MINUS 2 LOG LIKELIHOOD = -118.0541 WGMS = 0.0172 STD.ERROR=SQRT(WGMS) = 0.1312 ITERATION 26 BOUNDARIES: 0.171 0.588 MEANS: 0.062 0.378 0.733 WGSS = 17.2507 MINUS 2 LOG LIKELIHOOD = -118.6952 WGMS = 0.0173 STD.ERROR=SQRT(WGMS) = 0.1316 ITERATION 27 BOUNDARIES: 0.168 0.587 MEANS: 0.061 0.376 0.732 WGSS = 17.3404 MINUS 2 LOG LIKELIHOOD = -119.3219 WGMS = 0.0174 STD.ERROR=SQRT(WGMS) = 0.1319 ITERATION 28 BOUNDARIES: 0.165 0.586 MEANS: 0.060 0.375 0.732 WGSS = 17.4248 MINUS 2 LOG LIKELIHOOD = -119.9408 WGMS = 0.0175 STD.ERROR=SQRT(WGMS) = 0.1323 ITERATION 29 BOUNDARIES: 0.163 0.585 MEANS: 0.060 0.373 0.731 WGSS = 17.5045 MINUS 2 LOG LIKELIHOOD = -120.5560 WGMS = 0.0176 STD.ERROR=SQRT(WGMS) = 0.1326 ITERATION 30 BOUNDARIES: 0.160 0.584 MEANS: 0.059 0.371 0.731 WGSS = 17.5801 MINUS 2 LOG LIKELIHOOD = -121.1705 WGMS = 0.0177 STD.ERROR=SQRT(WGMS) = 0.1329 ITERATION 31 BOUNDARIES: 0.158 0.583 MEANS: 0.058 0.370 0.730 WGSS = 17.6521 MINUS 2 LOG LIKELIHOOD = -121.7892 WGMS = 0.0177 STD.ERROR=SQRT(WGMS) = 0.1331 ITERATION 32 BOUNDARIES: 0.156 0.582 MEANS: 0.057 0.368 0.730 WGSS = 17.7208 MINUS 2 LOG LIKELIHOOD = -122.4152 WGMS = 0.0178 STD.ERROR=SQRT(WGMS) = 0.1334 ITERATION 33 BOUNDARIES: 0.153 0.581 MEANS: 0.056 0.367 0.729 WGSS = 17.7866 MINUS 2 LOG LIKELIHOOD = -123.0507 WGMS = 0.0179 STD.ERROR=SQRT(WGMS) = 0.1336 ITERATION 34 BOUNDARIES: 0.151 0.579 MEANS: 0.055 0.365 0.729 WGSS = 17.8499 MINUS 2 LOG LIKELIHOOD = -123.6978 WGMS = 0.0179 STD.ERROR=SQRT(WGMS) = 0.1339 ITERATION 35 BOUNDARIES: 0.148 0.578 MEANS: 0.054 0.363 0.728 WGSS = 17.9108 MINUS 2 LOG LIKELIHOOD = -124.3584 WGMS = 0.0180 STD.ERROR=SQRT(WGMS) = 0.1341 ITERATION 36 BOUNDARIES: 0.146 0.577 MEANS: 0.053 0.362 0.728 WGSS = 17.9695 MINUS 2 LOG LIKELIHOOD = -125.0324 WGMS = 0.0180 STD.ERROR=SQRT(WGMS) = 0.1343 ITERATION 37 BOUNDARIES: 0.143 0.576 MEANS: 0.053 0.360 0.727 WGSS = 18.0259 MINUS 2 LOG LIKELIHOOD = -125.7177 WGMS = 0.0181 STD.ERROR=SQRT(WGMS) = 0.1345 ITERATION 38 BOUNDARIES: 0.141 0.575 MEANS: 0.052 0.359 0.727 WGSS = 18.0799 MINUS 2 LOG LIKELIHOOD = -126.4115 WGMS = 0.0182 STD.ERROR=SQRT(WGMS) = 0.1347 ITERATION 39 BOUNDARIES: 0.139 0.573 MEANS: 0.051 0.357 0.726 WGSS = 18.1314 MINUS 2 LOG LIKELIHOOD = -127.1084 WGMS = 0.0182 STD.ERROR=SQRT(WGMS) = 0.1349 ITERATION 40 BOUNDARIES: 0.136 0.572 MEANS: 0.050 0.355 0.726 WGSS = 18.1800 MINUS 2 LOG LIKELIHOOD = -127.8020 WGMS = 0.0183 STD.ERROR=SQRT(WGMS) = 0.1351 ITERATION 41 BOUNDARIES: 0.134 0.571 MEANS: 0.049 0.354 0.725 WGSS = 18.2254 MINUS 2 LOG LIKELIHOOD = -128.4828 WGMS = 0.0183 STD.ERROR=SQRT(WGMS) = 0.1353 ITERATION 42 BOUNDARIES: 0.132 0.570 MEANS: 0.048 0.352 0.724 WGSS = 18.2670 MINUS 2 LOG LIKELIHOOD = -129.1424 WGMS = 0.0183 STD.ERROR=SQRT(WGMS) = 0.1354 ITERATION 43 BOUNDARIES: 0.129 0.569 MEANS: 0.048 0.351 0.724 WGSS = 18.3045 MINUS 2 LOG LIKELIHOOD = -129.7700 WGMS = 0.0184 STD.ERROR=SQRT(WGMS) = 0.1356 ITERATION 44 BOUNDARIES: 0.127 0.567 MEANS: 0.047 0.349 0.723 WGSS = 18.3373 MINUS 2 LOG LIKELIHOOD = -130.3587 WGMS = 0.0184 STD.ERROR=SQRT(WGMS) = 0.1357 ITERATION 45 BOUNDARIES: 0.125 0.566 MEANS: 0.046 0.348 0.723 WGSS = 18.3649 MINUS 2 LOG LIKELIHOOD = -130.9006 WGMS = 0.0184 STD.ERROR=SQRT(WGMS) = 0.1358 ITERATION 46 BOUNDARIES: 0.123 0.565 MEANS: 0.046 0.346 0.722 WGSS = 18.3871 MINUS 2 LOG LIKELIHOOD = -131.3924 WGMS = 0.0185 STD.ERROR=SQRT(WGMS) = 0.1359 ITERATION 47 BOUNDARIES: 0.121 0.564 MEANS: 0.045 0.345 0.721 WGSS = 18.4037 MINUS 2 LOG LIKELIHOOD = -131.8340 WGMS = 0.0185 STD.ERROR=SQRT(WGMS) = 0.1359 ITERATION 48 BOUNDARIES: 0.119 0.562 MEANS: 0.045 0.344 0.721 WGSS = 18.4145 MINUS 2 LOG LIKELIHOOD = -132.2252 WGMS = 0.0185 STD.ERROR=SQRT(WGMS) = 0.1360 ITERATION 49 BOUNDARIES: 0.118 0.561 MEANS: 0.044 0.342 0.720 NO CASE CHANGED CLUSTERS IN THIS ITERATION. WGSS = 18.4196 MINUS 2 LOG LIKELIHOOD = -132.5685 WGMS = 0.0185 STD.ERROR=SQRT(WGMS) = 0.1360 ITERATION 50 BOUNDARIES: 0.116 0.560 MEANS: 0.044 0.341 0.720 WGSS = 18.4191 MINUS 2 LOG LIKELIHOOD = -132.8701 WGMS = 0.0185 STD.ERROR=SQRT(WGMS) = 0.1360 ITERATION 51 BOUNDARIES: 0.115 0.559 MEANS: 0.043 0.340 0.719 WGSS = 18.4134 MINUS 2 LOG LIKELIHOOD = -133.1332 WGMS = 0.0185 STD.ERROR=SQRT(WGMS) = 0.1360 ITERATION 52 BOUNDARIES: 0.113 0.558 MEANS: 0.043 0.339 0.718 WGSS = 18.4028 MINUS 2 LOG LIKELIHOOD = -133.3627 WGMS = 0.0185 STD.ERROR=SQRT(WGMS) = 0.1359 ITERATION 53 BOUNDARIES: 0.112 0.556 MEANS: 0.043 0.338 0.718 NO CASE CHANGED CLUSTERS IN THIS ITERATION. PROBS: 0.156 0.543 0.301 NUMBERS: 174 515 310 VARIANCES: 0.001 0.027 0.012 STD.DEVS.: 0.028 0.165 0.107 M.L. ESTIMATE OF COMMON VARIANCE = 0.01842 NUMBER OF PARAMETERS = 8 AIC = -117.3627 SCHWARZ CRITERION = -78.1087 KASHYAP CRITERION = -66.8191