The SAS System 1 07:12 Thursday, April 5, 2012 The GENMOD Procedure Model Information Data Set WORK.SHIPS Distribution Poisson Link Function Log Dependent Variable dmgincd Offset Variable lnaggsrvc Number of Observations Read 34 Number of Observations Used 34 Class Level Information Class Levels Values shiptype 5 1 2 3 4 5 constprd 4 1 2 3 4 operprd 2 1 2 Criteria For Assessing Goodness Of Fit Criterion DF Value Value/DF Deviance 26 49.3552 1.8983 Scaled Deviance 26 49.3552 1.8983 Pearson Chi-Square 26 51.3422 1.9747 Scaled Pearson X2 26 51.3422 1.9747 Log Likelihood 762.6831 Full Log Likelihood -73.6108 AIC (smaller is better) 163.2217 AICC (smaller is better) 168.9817 BIC (smaller is better) 175.4326 Algorithm converged. Analysis Of Maximum Likelihood Parameter Estimates Standard Wald 95% Confidence Wald Parameter DF Estimate Error Limits Chi-Square Intercept 1 -5.2564 0.2475 -5.7416 -4.7712 450.92 shiptype 1 1 -0.3042 0.2358 -0.7664 0.1579 1.66 shiptype 2 1 -0.8629 0.1982 -1.2514 -0.4743 18.94 shiptype 3 1 -0.9881 0.3393 -1.6532 -0.3230 8.48 The SAS System 2 07:12 Thursday, April 5, 2012 The GENMOD Procedure Analysis Of Maximum Likelihood Parameter Estimates Standard Wald 95% Confidence Wald Parameter DF Estimate Error Limits Chi-Square shiptype 4 1 -0.3746 0.3073 -0.9769 0.2278 1.49 shiptype 5 0 0.0000 0.0000 0.0000 0.0000 . constprd 1 1 -0.7083 0.2215 -1.1425 -0.2741 10.22 constprd 2 1 0.0452 0.2015 -0.3497 0.4401 0.05 constprd 3 1 0.2570 0.1976 -0.1303 0.6443 1.69 constprd 4 0 0.0000 0.0000 0.0000 0.0000 . Scale 0 1.0000 0.0000 1.0000 1.0000 Analysis Of Maximum Likelihood Parameter Estimates Parameter Pr > ChiSq Intercept <.0001 shiptype 1 0.1970 shiptype 2 <.0001 shiptype 3 0.0036 shiptype 4 0.2229 shiptype 5 . constprd 1 0.0014 constprd 2 0.8225 constprd 3 0.1933 constprd 4 . Scale NOTE: The scale parameter was held fixed. The SAS System 3 07:12 Thursday, April 5, 2012 The GENMOD Procedure Model Information Data Set WORK.SHIPS Distribution Poisson Link Function Log Dependent Variable dmgincd Offset Variable lnaggsrvc Number of Observations Read 34 Number of Observations Used 34 Class Level Information Class Levels Values shiptype 5 1 2 3 4 5 constprd 4 1 2 3 4 operprd 2 1 2 Criteria For Assessing Goodness Of Fit Criterion DF Value Value/DF Deviance 28 70.1029 2.5037 Scaled Deviance 28 70.1029 2.5037 Pearson Chi-Square 28 77.9367 2.7835 Scaled Pearson X2 28 77.9367 2.7835 Log Likelihood 752.3092 Full Log Likelihood -83.9847 AIC (smaller is better) 179.9694 AICC (smaller is better) 183.0805 BIC (smaller is better) 189.1276 Algorithm converged. Analysis Of Maximum Likelihood Parameter Estimates Standard Wald 95% Confidence Wald Parameter DF Estimate Error Limits Chi-Square Intercept 1 -4.9108 0.1795 -5.2625 -4.5591 748.82 shiptype 1 1 -0.3885 0.2348 -0.8487 0.0717 2.74 shiptype 2 1 -1.1315 0.1887 -1.5013 -0.7616 35.95 shiptype 3 1 -1.1430 0.3385 -1.8065 -0.4795 11.40 The SAS System 4 07:12 Thursday, April 5, 2012 The GENMOD Procedure Analysis Of Maximum Likelihood Parameter Estimates Standard Wald 95% Confidence Wald Parameter DF Estimate Error Limits Chi-Square shiptype 4 1 -0.5730 0.3006 -1.1621 0.0161 3.63 shiptype 5 0 0.0000 0.0000 0.0000 0.0000 . operprd 1 1 -0.5030 0.1115 -0.7216 -0.2845 20.35 operprd 2 0 0.0000 0.0000 0.0000 0.0000 . Scale 0 1.0000 0.0000 1.0000 1.0000 Analysis Of Maximum Likelihood Parameter Estimates Parameter Pr > ChiSq Intercept <.0001 shiptype 1 0.0980 shiptype 2 <.0001 shiptype 3 0.0007 shiptype 4 0.0566 shiptype 5 . operprd 1 <.0001 operprd 2 . Scale NOTE: The scale parameter was held fixed. The SAS System 5 07:12 Thursday, April 5, 2012 The GENMOD Procedure Model Information Data Set WORK.SHIPS Distribution Poisson Link Function Log Dependent Variable dmgincd Offset Variable lnaggsrvc Number of Observations Read 34 Number of Observations Used 34 Class Level Information Class Levels Values shiptype 5 1 2 3 4 5 constprd 4 1 2 3 4 operprd 2 1 2 Criteria For Assessing Goodness Of Fit Criterion DF Value Value/DF Deviance 29 62.3653 2.1505 Scaled Deviance 29 62.3653 2.1505 Pearson Chi-Square 29 82.7371 2.8530 Scaled Pearson X2 29 82.7371 2.8530 Log Likelihood 756.1780 Full Log Likelihood -80.1159 AIC (smaller is better) 170.2318 AICC (smaller is better) 172.3747 BIC (smaller is better) 177.8636 Algorithm converged. Analysis Of Maximum Likelihood Parameter Estimates Standard Wald 95% Confidence Wald Parameter DF Estimate Error Limits Chi-Square Intercept 1 -5.8561 0.1690 -6.1874 -5.5248 1200.30 constprd 1 1 -0.7041 0.2203 -1.1359 -0.2723 10.21 constprd 2 1 0.0502 0.1984 -0.3387 0.4390 0.06 constprd 3 1 0.3468 0.1948 -0.0349 0.7286 3.17 The SAS System 6 07:12 Thursday, April 5, 2012 The GENMOD Procedure Analysis Of Maximum Likelihood Parameter Estimates Standard Wald 95% Confidence Wald Parameter DF Estimate Error Limits Chi-Square constprd 4 0 0.0000 0.0000 0.0000 0.0000 . operprd 1 1 -0.3875 0.1181 -0.6189 -0.1560 10.76 operprd 2 0 0.0000 0.0000 0.0000 0.0000 . Scale 0 1.0000 0.0000 1.0000 1.0000 Analysis Of Maximum Likelihood Parameter Estimates Parameter Pr > ChiSq Intercept <.0001 constprd 1 0.0014 constprd 2 0.8005 constprd 3 0.0750 constprd 4 . operprd 1 0.0010 operprd 2 . Scale NOTE: The scale parameter was held fixed. The SAS System 7 07:12 Thursday, April 5, 2012 The GENMOD Procedure Model Information Data Set WORK.SHIPS Distribution Poisson Link Function Log Dependent Variable dmgincd Offset Variable lnaggsrvc Number of Observations Read 34 Number of Observations Used 34 Class Level Information Class Levels Values shiptype 5 1 2 3 4 5 constprd 4 1 2 3 4 operprd 2 1 2 Criteria For Assessing Goodness Of Fit Criterion DF Value Value/DF Deviance 25 38.6951 1.5478 Scaled Deviance 25 38.6951 1.5478 Pearson Chi-Square 25 42.2753 1.6910 Scaled Pearson X2 25 42.2753 1.6910 Log Likelihood 768.0131 Full Log Likelihood -68.2808 AIC (smaller is better) 154.5615 AICC (smaller is better) 162.0615 BIC (smaller is better) 168.2988 Algorithm converged. Analysis Of Maximum Likelihood Parameter Estimates Standard Wald 95% Confidence Wald Parameter DF Estimate Error Limits Chi-Square Intercept 1 -5.2424 0.2473 -5.7271 -4.7577 449.36 shiptype 1 1 -0.3256 0.2359 -0.7879 0.1367 1.91 shiptype 2 1 -0.8689 0.1984 -1.2578 -0.4800 19.18 shiptype 3 1 -1.0130 0.3395 -1.6783 -0.3477 8.91 The SAS System 8 07:12 Thursday, April 5, 2012 The GENMOD Procedure Analysis Of Maximum Likelihood Parameter Estimates Standard Wald 95% Confidence Wald Parameter DF Estimate Error Limits Chi-Square shiptype 4 1 -0.4015 0.3071 -1.0035 0.2004 1.71 shiptype 5 0 0.0000 0.0000 0.0000 0.0000 . constprd 1 1 -0.4534 0.2332 -0.9104 0.0036 3.78 constprd 2 1 0.2437 0.2088 -0.1654 0.6529 1.36 constprd 3 1 0.3650 0.1995 -0.0259 0.7559 3.35 constprd 4 0 0.0000 0.0000 0.0000 0.0000 . operprd 1 1 -0.3845 0.1183 -0.6163 -0.1527 10.57 operprd 2 0 0.0000 0.0000 0.0000 0.0000 . Scale 0 1.0000 0.0000 1.0000 1.0000 Analysis Of Maximum Likelihood Parameter Estimates Parameter Pr > ChiSq Intercept <.0001 shiptype 1 0.1675 shiptype 2 <.0001 shiptype 3 0.0028 shiptype 4 0.1911 shiptype 5 . constprd 1 0.0518 constprd 2 0.2430 constprd 3 0.0673 constprd 4 . operprd 1 0.0012 operprd 2 . Scale NOTE: The scale parameter was held fixed. LR Statistics For Type 3 Analysis Chi- Source DF Square Pr > ChiSq shiptype 4 23.67 <.0001 constprd 3 31.41 <.0001 operprd 1 10.66 0.0011