The SAS System 1 10:15 Tuesday, April 8, 2008 The GENMOD Procedure Model Information Data Set WORK.ONE Distribution Binomial Link Function Logit Response Variable (Events) casualty Response Variable (Trials) totsoldier Number of Observations Read 18 Number of Observations Used 18 Number of Events 23045 Number of Trials 95369 Class Level Information Class Levels Values division 9 1 2 3 4 5 6 7 8 9 Criteria For Assessing Goodness Of Fit Criterion DF Value Value/DF Deviance 17 18665.9088 1097.9946 Scaled Deviance 17 18665.9088 1097.9946 Pearson Chi-Square 17 17167.3854 1009.8462 Scaled Pearson X2 17 17167.3854 1009.8462 Log Likelihood -52735.5645 Algorithm converged. Analysis Of Parameter Estimates Standard Wald 95% Chi- Parameter DF Estimate Error Confidence Limits Square Pr > ChiSq Intercept 1 -1.1437 0.0076 -1.1585 -1.1289 22860.3 <.0001 Scale 0 1.0000 0.0000 1.0000 1.0000 NOTE: The scale parameter was held fixed. The SAS System 2 10:15 Tuesday, April 8, 2008 The GENMOD Procedure Model Information Data Set WORK.ONE Distribution Binomial Link Function Logit Response Variable (Events) casualty Response Variable (Trials) totsoldier Number of Observations Read 18 Number of Observations Used 18 Number of Events 23045 Number of Trials 95369 Class Level Information Class Levels Values division 9 1 2 3 4 5 6 7 8 9 Criteria For Assessing Goodness Of Fit Criterion DF Value Value/DF Deviance 16 18655.1812 1165.9488 Scaled Deviance 16 18655.1812 1165.9488 Pearson Chi-Square 16 17165.3130 1072.8321 Scaled Pearson X2 16 17165.3130 1072.8321 Log Likelihood -52730.2007 Algorithm converged. Analysis Of Parameter Estimates Standard Wald 95% Chi- Parameter DF Estimate Error Confidence Limits Square Pr > ChiSq Intercept 1 -1.1502 0.0078 -1.1656 -1.1349 21578.1 <.0001 officer 1 0.1001 0.0303 0.0406 0.1595 10.88 0.0010 Scale 0 1.0000 0.0000 1.0000 1.0000 NOTE: The scale parameter was held fixed. The SAS System 3 10:15 Tuesday, April 8, 2008 The GENMOD Procedure Model Information Data Set WORK.ONE Distribution Binomial Link Function Logit Response Variable (Events) casualty Response Variable (Trials) totsoldier Number of Observations Read 18 Number of Observations Used 18 Number of Events 23045 Number of Trials 95369 Class Level Information Class Levels Values division 9 1 2 3 4 5 6 7 8 9 Criteria For Assessing Goodness Of Fit Criterion DF Value Value/DF Deviance 8 28.5736 3.5717 Scaled Deviance 8 28.5736 3.5717 Pearson Chi-Square 8 30.0673 3.7584 Scaled Pearson X2 8 30.0673 3.7584 Log Likelihood -43416.8969 Algorithm converged. Analysis Of Parameter Estimates Standard Wald 95% Confidence Chi- Parameter DF Estimate Error Limits Square Intercept 1 -2.2553 0.0676 -2.3878 -2.1229 1113.53 officer 1 0.0495 0.0336 -0.0163 0.1153 2.18 division 1 1 2.6764 0.0706 2.5380 2.8148 1437.01 division 2 1 1.5844 0.0701 1.4470 1.7218 510.85 division 3 1 1.6467 0.0702 1.5091 1.7844 549.55 division 4 1 0.7004 0.0716 0.5601 0.8406 95.78 division 5 1 -1.8955 0.0936 -2.0790 -1.7120 409.92 division 6 1 1.7897 0.0707 1.6512 1.9282 641.19 The SAS System 4 10:15 Tuesday, April 8, 2008 The GENMOD Procedure Analysis Of Parameter Estimates Standard Wald 95% Confidence Chi- Parameter DF Estimate Error Limits Square division 7 1 0.3152 0.0750 0.1682 0.4622 17.66 division 8 1 -0.2729 0.0764 -0.4226 -0.1232 12.76 division 9 0 0.0000 0.0000 0.0000 0.0000 . Scale 0 1.0000 0.0000 1.0000 1.0000 Analysis Of Parameter Estimates Parameter Pr > ChiSq Intercept <.0001 officer 0.1400 division 1 <.0001 division 2 <.0001 division 3 <.0001 division 4 <.0001 division 5 <.0001 division 6 <.0001 division 7 <.0001 division 8 0.0004 division 9 . Scale NOTE: The scale parameter was held fixed. The SAS System 5 10:15 Tuesday, April 8, 2008 The GENMOD Procedure Model Information Data Set WORK.ONE Distribution Binomial Link Function Logit Response Variable (Events) casualty Response Variable (Trials) totsoldier Number of Observations Read 18 Number of Observations Used 18 Number of Events 23045 Number of Trials 95369 Class Level Information Class Levels Values division 9 1 2 3 4 5 6 7 8 9 Criteria For Assessing Goodness Of Fit Criterion DF Value Value/DF Deviance 0 0.0000 . Scaled Deviance 0 0.0000 . Pearson Chi-Square 0 0.0000 . Scaled Pearson X2 0 0.0000 . Log Likelihood -43402.6101 Algorithm converged. Analysis Of Parameter Estimates Standard Wald 95% Chi- Parameter DF Estimate Error Confidence Limits Square Intercept 1 -2.2919 0.0700 -2.4290 -2.1548 1073.35 division 1 1 2.7349 0.0731 2.5916 2.8782 1399.46 division 2 1 1.6134 0.0726 1.4711 1.7556 494.05 division 3 1 1.6759 0.0727 1.5334 1.8185 531.05 division 4 1 0.7299 0.0741 0.5846 0.8751 97.03 division 5 1 -1.8563 0.0969 -2.0462 -1.6664 367.04 division 6 1 1.8300 0.0731 1.6867 1.9733 626.17 division 7 1 0.3595 0.0776 0.2075 0.5115 21.48 The SAS System 6 10:15 Tuesday, April 8, 2008 The GENMOD Procedure Analysis Of Parameter Estimates Standard Wald 95% Chi- Parameter DF Estimate Error Confidence Limits Square division 8 1 -0.2452 0.0790 -0.4001 -0.0903 9.63 division 9 0 0.0000 0.0000 0.0000 0.0000 . officer 1 0.7684 0.2767 0.2261 1.3106 7.71 officer*division 1 1 -1.0244 0.2876 -1.5880 -0.4608 12.69 officer*division 2 1 -0.6183 0.2854 -1.1777 -0.0589 4.69 officer*division 3 1 -0.6145 0.2865 -1.1761 -0.0529 4.60 officer*division 4 1 -0.6161 0.2916 -1.1876 -0.0447 4.47 officer*division 5 1 -0.7583 0.3802 -1.5035 -0.0132 3.98 officer*division 6 1 -0.7852 0.2907 -1.3550 -0.2154 7.29 officer*division 7 1 -0.8456 0.3093 -1.4518 -0.2394 7.47 officer*division 8 1 -0.5623 0.3146 -1.1790 0.0543 3.19 officer*division 9 0 0.0000 0.0000 0.0000 0.0000 . Scale 0 1.0000 0.0000 1.0000 1.0000 Analysis Of Parameter Estimates Parameter Pr > ChiSq Intercept <.0001 division 1 <.0001 division 2 <.0001 division 3 <.0001 division 4 <.0001 division 5 <.0001 division 6 <.0001 division 7 <.0001 division 8 0.0019 division 9 . officer 0.0055 officer*division 1 0.0004 officer*division 2 0.0303 officer*division 3 0.0320 officer*division 4 0.0346 officer*division 5 0.0461 officer*division 6 0.0069 officer*division 7 0.0063 officer*division 8 0.0739 officer*division 9 . Scale NOTE: The scale parameter was held fixed.