Method = 1 (Strong Hierarchy) was chosen. The program 1. Standardizes all predictors to have mean 0 and variance 1 and renames them as x1-xp (see table below), numbered as they occur in the data set. 2. Fits main effects, interactions, and squared terms according to a modified forward selection procedure that on average has False Selection Rate (FSR) = 0.05. Developed by Hugh Crews, August 2008. The number of effects forced into the model is 0. Note: The terms=k option was used. Suboptimal results may occur if k is so small that gammahat never reaches its maximum. Variable names Variable Obs name Original name 1 x1 duration 2 x2 amount 3 x3 rate 4 x4 residence 5 x5 age 6 x6 exist_cr 7 x7 provide 8 x8 nchecking 9 x9 nhistory 10 x10 nsavings 11 x11 nemploy 12 x12 njob 13 x13 nphone 14 x14 nforeign 15 x15 debtors_1 16 x16 debtors_2 17 x17 housing_1 18 x18 housing_2 19 x19 other_plans_1 20 x20 other_plans_2 21 x21 property_1 22 x22 property_2 23 x23 property_3 24 x24 purpose_1 25 x25 purpose_2 26 x26 purpose_3 27 x27 purpose_4 28 x28 purpose_5 29 x29 purpose_6 30 x30 purpose_7 31 x31 purpose_8 32 x32 purpose_9 33 x33 status_1 34 x34 status_2 35 x35 status_3 36 y goodbad Forward Sequence Selection Effect p-to- monotonized Obs Step 1+S Entered enter p-to-enter 1 0 1 Intercept <.0001 0.000000 2 1 2 x9 <.0001 0.000000 3 2 3 x1 <.0001 0.000000 4 3 4 x24 <.0001 0.000027 5 4 5 x31 0.0009 0.000915 6 5 6 x35 0.0016 0.001629 7 6 7 x25 0.0057 0.005691 8 7 8 x18 0.0020 0.005691 9 8 9 x10 0.0055 0.005691 10 9 10 x10*x31 0.0066 0.006588 11 10 11 x1*x1 0.0149 0.014897 12 11 12 x18*x31 0.0184 0.018421 13 12 13 x19 0.0273 0.027286 14 13 14 x19*x24 0.0063 0.027286 15 14 15 x3 0.0230 0.027286 16 15 16 x2 0.0056 0.027286 17 16 17 x2*x25 0.0084 0.027286 18 17 18 x1*x2 0.0195 0.027286 19 18 19 x2*x2 0.0040 0.027286 20 19 20 x28 0.0156 0.027286 21 20 21 x5 0.0263 0.027286 22 21 22 x11 0.0214 0.027286 23 22 23 x14 0.0264 0.027286 24 23 24 x11*x11 0.0275 0.027471 25 24 25 x14*x35 0.0276 0.027636 26 25 26 x5*x24 0.0232 0.027636 27 26 27 x6 0.0356 0.035573 28 27 28 x6*x10 0.0002 0.035573 29 28 29 x1*x9 0.0209 0.035573 30 29 30 x6*x18 0.0203 0.035573 31 30 31 x9*x18 0.0204 0.035573 32 31 32 x13 0.0456 0.045644 33 32 33 x2*x13 0.0060 0.045644 34 33 34 x10*x10 0.0376 0.045644 35 34 35 x5*x19 0.0421 0.045644 36 35 36 x1*x13 0.0541 0.054125 37 36 37 x11*x28 0.0711 0.071082 38 37 38 x20 0.0670 0.071082 39 38 39 x5*x20 0.0062 0.071082 40 39 40 x3*x28 0.0676 0.071082 41 40 41 x33 0.0599 0.071082 Sequence of Possible Models Obs alpha gammahat 1+S 1 0.000000 0.00000 1 2 0.000000 0.00000 2 3 0.000000 0.00000 3 4 0.000027 0.00023 4 5 0.000915 0.00659 5 6 0.001629 0.01032 6 7 0.005691 0.03162 9 8 0.006588 0.03755 10 9 0.014897 0.07584 11 10 0.018421 0.08443 12 11 0.027286 0.13168 23 12 0.027471 0.14193 24 13 0.027636 0.12968 26 14 0.035573 0.15262 31 15 0.045644 0.18910 35 16 0.054125 0.21650 36 17 0.071082 0.27046 41 Fast FSR estimates alpha Obs estimate 1+S 1 .008771930 10 The LOGISTIC Procedure Model Information Data Set WORK.CREDIT_NEW Response Variable y goodbad Number of Response Levels 2 Model binary logit Optimization Technique Fisher's scoring Number of Observations Read 1000 Number of Observations Used 1000 Response Profile Ordered Total Value y Frequency 1 0 700 2 1 300 Probability modeled is y=0. Model Convergence Status Convergence criterion (GCONV=1E-8) satisfied. Model Fit Statistics Intercept Intercept and Criterion Only Covariates AIC 1223.729 1081.230 SC 1228.636 1130.307 -2 Log L 1221.729 1061.230 Testing Global Null Hypothesis: BETA=0 Test Chi-Square DF Pr > ChiSq Likelihood Ratio 160.4988 9 <.0001 Score 150.3956 9 <.0001 Wald 125.1838 9 <.0001 Analysis of Maximum Likelihood Estimates Standard Wald Parameter DF Estimate Error Chi-Square Pr > ChiSq Intercept 1 1.0047 0.0798 158.5912 <.0001 x9 1 0.5010 0.0820 37.2915 <.0001 x1 1 -0.5387 0.0769 49.1282 <.0001 x24 1 -0.2893 0.0756 14.6479 0.0001 x31 1 -0.1985 0.0736 7.2619 0.0070 x35 1 0.2245 0.0766 8.5842 0.0034 x25 1 0.2953 0.0921 10.2719 0.0014 x18 1 0.2189 0.0746 8.5972 0.0034 x10 1 0.2696 0.0984 7.5095 0.0061 x31*x10 1 -0.2097 0.0934 5.0380 0.0248 Odds Ratio Estimates Point 95% Wald Effect Estimate Confidence Limits x9 1.650 1.405 1.938 x1 0.583 0.502 0.678 x24 0.749 0.646 0.868 x35 1.252 1.077 1.454 x25 1.344 1.122 1.609 x18 1.245 1.075 1.441 Association of Predicted Probabilities and Observed Responses Percent Concordant 74.3 Somers' D 0.490 Percent Discordant 25.3 Gamma 0.493 Percent Tied 0.4 Tau-a 0.206 Pairs 210000 c 0.745