Final Generalized Gamma Model 1 The NLMIXED Procedure Specifications Data Set WORK.HIV Dependent Variable exit Distribution for Dependent Variable General Optimization Technique Dual Quasi-Newton Integration Method None Final Generalized Gamma Model 2 ----------------------------------- period=1 ----------------------------------- The NLMIXED Procedure Dimensions Observations Used 633 Observations Not Used 0 Total Observations 633 Parameters 2 Parameters beta sigma NegLogLike 0 1 865.296989 Iteration History Iter Calls NegLogLike Diff MaxGrad Slope 1 6 808.559901 56.73709 81.67648 -2014.32 2 10 794.738108 13.82179 64.39231 -79.7532 3 16 785.781216 8.956892 40.61399 -9.72195 4 20 784.161789 1.619427 3.260004 -7.01258 5 23 784.131748 0.030041 0.935949 -0.06071 6 27 784.13085 0.000898 0.0168 -0.0015 7 31 784.13085 3.82E-7 0.000117 -7.86E-7 NOTE: GCONV convergence criterion satisfied. Fit Statistics -2 Log Likelihood 1568.3 AIC (smaller is better) 1572.3 AICC (smaller is better) 1572.3 BIC (smaller is better) 1581.2 Parameter Estimates Standard Parameter Estimate Error DF t Value Pr > |t| Alpha Lower beta 0.6365 0.05092 633 12.50 <.0001 0.05 0.5365 sigma 0.7978 0.06052 633 13.18 <.0001 0.05 0.6789 Final Generalized Gamma Model 3 ----------------------------------- period=1 ----------------------------------- The NLMIXED Procedure Parameter Estimates Parameter Upper Gradient beta 0.7365 -0.00006 sigma 0.9166 -0.00012 Covariance Matrix of Parameter Estimates Row Parameter beta sigma 1 beta 0.002593 -0.00171 2 sigma -0.00171 0.003663 Additional Estimates Standard Label Estimate Error DF t Value Pr > |t| Alpha Lower logsigma -0.2259 0.07586 633 -2.98 0.0030 0.05 -0.3749 Additional Estimates Label Upper logsigma -0.07696 Final Generalized Gamma Model 4 ----------------------------------- period=2 ----------------------------------- The NLMIXED Procedure Dimensions Observations Used 660 Observations Not Used 0 Total Observations 660 Parameters 2 Parameters beta sigma NegLogLike 0 1 964.773941 Iteration History Iter Calls NegLogLike Diff MaxGrad Slope 1 6 818.396044 146.3779 150.73 -2363.41 2 10 813.005919 5.390125 213.4897 -395.175 3 13 804.136218 8.869701 68.91261 -27.3924 4 16 802.272007 1.864211 17.70168 -3.13992 5 19 802.162997 0.10901 2.169268 -0.19578 6 23 802.161301 0.001695 0.044984 -0.00329 7 27 802.161301 7.363E-7 0.000184 -1.52E-6 NOTE: GCONV convergence criterion satisfied. Fit Statistics -2 Log Likelihood 1604.3 AIC (smaller is better) 1608.3 AICC (smaller is better) 1608.3 BIC (smaller is better) 1617.3 Parameter Estimates Standard Parameter Estimate Error DF t Value Pr > |t| Alpha Lower beta 0.6483 0.04070 660 15.93 <.0001 0.05 0.5684 sigma 0.8483 0.02435 660 34.85 <.0001 0.05 0.8005 Final Generalized Gamma Model 5 ----------------------------------- period=2 ----------------------------------- The NLMIXED Procedure Parameter Estimates Parameter Upper Gradient beta 0.7282 0.000184 sigma 0.8962 0.000047 Covariance Matrix of Parameter Estimates Row Parameter beta sigma 1 beta 0.001656 0.000255 2 sigma 0.000255 0.000593 Additional Estimates Standard Label Estimate Error DF t Value Pr > |t| Alpha Lower logsigma -0.1645 0.02870 660 -5.73 <.0001 0.05 -0.2208 Additional Estimates Label Upper logsigma -0.1081 Final Generalized Gamma Model 6 ----------------------------------- period=3 ----------------------------------- The NLMIXED Procedure Dimensions Observations Used 472 Observations Not Used 0 Total Observations 472 Parameters 2 Parameters beta sigma NegLogLike 0 1 836.996536 Iteration History Iter Calls NegLogLike Diff MaxGrad Slope 1 6 431.328381 405.6682 94.88909 -13469.3 2 13 362.758581 68.5698 10.97413 -40.9358 3 17 362.434707 0.323874 9.130243 -1.2306 4 26 358.626114 3.808593 2.680523 -0.13543 5 32 358.518261 0.107853 0.318878 -0.11938 6 36 358.514848 0.003413 0.039515 -0.00522 7 40 358.514823 0.000026 0.001418 -0.00005 8 44 358.514823 9.004E-8 0.000013 -1.66E-7 NOTE: GCONV convergence criterion satisfied. Fit Statistics -2 Log Likelihood 717.0 AIC (smaller is better) 721.0 AICC (smaller is better) 721.1 BIC (smaller is better) 729.3 Parameter Estimates Standard Parameter Estimate Error DF t Value Pr > |t| Alpha Lower beta 2.4320 0.2115 472 11.50 <.0001 0.05 2.0163 sigma 2.0824 0.3004 472 6.93 <.0001 0.05 1.4921 Final Generalized Gamma Model 7 ----------------------------------- period=3 ----------------------------------- The NLMIXED Procedure Parameter Estimates Parameter Upper Gradient beta 2.8477 0.000013 sigma 2.6727 -1.96E-6 Covariance Matrix of Parameter Estimates Row Parameter beta sigma 1 beta 0.04475 0.04019 2 sigma 0.04019 0.09023 Additional Estimates Standard Label Estimate Error DF t Value Pr > |t| Alpha Lower logsigma 0.7335 0.1443 472 5.08 <.0001 0.05 0.4501 Additional Estimates Label Upper logsigma 1.0170 Final Generalized Gamma Model 8 ----------------------------------- period=4 ----------------------------------- The NLMIXED Procedure Dimensions Observations Used 549 Observations Not Used 0 Total Observations 549 Parameters 2 Parameters beta sigma NegLogLike 0 1 1956.9227 Iteration History Iter Calls NegLogLike Diff MaxGrad Slope 1 7 338.474313 1618.448 28.02309 -238020 2 16 314.026678 24.44764 11.49141 -3.08548 3 19 311.358523 2.668155 4.771837 -12.4264 4 22 310.309661 1.048862 2.853149 -1.77509 5 25 310.278378 0.031283 2.968222 -0.06117 6 34 309.820576 0.457802 7.943653 -0.00713 7 40 308.44032 1.380256 3.405 -1.72174 8 43 307.223495 1.216825 12.37801 -2.27718 9 46 306.272261 0.951233 1.300486 -8.08116 10 50 306.245294 0.026967 0.436346 -0.03028 11 54 306.240014 0.005281 0.166203 -0.01906 12 58 306.239623 0.00039 0.054284 -0.00102 13 62 306.239598 0.000026 0.001302 -0.00005 14 66 306.239598 1.677E-8 0.00003 -3.05E-8 NOTE: GCONV convergence criterion satisfied. Fit Statistics -2 Log Likelihood 612.5 AIC (smaller is better) 616.5 AICC (smaller is better) 616.5 BIC (smaller is better) 625.1 Final Generalized Gamma Model 9 ----------------------------------- period=4 ----------------------------------- The NLMIXED Procedure Parameter Estimates Standard Parameter Estimate Error DF t Value Pr > |t| Alpha Lower beta 3.1061 0.1737 549 17.88 <.0001 0.05 2.7649 sigma 1.2925 0.2448 549 5.28 <.0001 0.05 0.8116 Parameter Estimates Parameter Upper Gradient beta 3.4473 0.00003 sigma 1.7734 0.000023 Covariance Matrix of Parameter Estimates Row Parameter beta sigma 1 beta 0.03018 0.03011 2 sigma 0.03011 0.05994 Additional Estimates Standard Label Estimate Error DF t Value Pr > |t| Alpha Lower logsigma 0.2566 0.1894 549 1.35 0.1761 0.05 -0.1155 Additional Estimates Label Upper logsigma 0.6287