The SAS System 15:44 Friday, May 9, 2008 1 The PHREG Procedure Model Information Data Set WORK.A Dependent Variable time Censoring Variable delta Censoring Value(s) 0 Ties Handling BRESLOW Number of Observations Read 312 Number of Observations Used 312 Summary of the Number of Event and Censored Values Percent Total Event Censored Censored 312 125 187 59.94 Convergence Status Convergence criterion (GCONV=1E-8) satisfied. Model Fit Statistics Without With Criterion Covariates Covariates -2 LOG L 1279.960 1080.825 AIC 1279.960 1090.825 SBC 1279.960 1104.966 Testing Global Null Hypothesis: BETA=0 Test Chi-Square DF Pr > ChiSq Likelihood Ratio 199.1349 5 <.0001 Score 269.3073 5 <.0001 Wald 198.2020 5 <.0001 The SAS System 15:44 Friday, May 9, 2008 2 The PHREG Procedure Analysis of Maximum Likelihood Estimates Parameter Standard Hazard Parameter DF Estimate Error Chi-Square Pr > ChiSq Ratio logbili 1 0.87921 0.09873 79.2980 <.0001 2.409 logalb 1 -3.05332 0.72408 17.7818 <.0001 0.047 age 1 0.03327 0.00866 14.7565 0.0001 1.034 logproth 1 3.01567 1.02380 8.6764 0.0032 20.403 edema 1 0.78470 0.29913 6.8814 0.0087 2.192 Deviance Residuals for Mayo Model 15:44 Friday, May 9, 2008 3 The PHREG Procedure Model Information Data Set WORK.A Dependent Variable time Censoring Variable delta Censoring Value(s) 0 Ties Handling BRESLOW Number of Observations Read 312 Number of Observations Used 312 Summary of the Number of Event and Censored Values Percent Total Event Censored Censored 312 125 187 59.94 Convergence Status WARNING: Iteration limit reached without convergence. WARNING: Convergence was not attained in 0 iterations. WARNING: The PHREG procedure continues in spite of the above warning. Results shown are based on the last maximum likelihood iteration. Validity of the model fit is questionable. Model Fit Statistics Without With Criterion Covariates Covariates -2 LOG L 1279.960 1279.960 AIC 1279.960 1289.960 SBC 1279.960 1304.101 Testing Global Null Hypothesis: BETA=0 Test Chi-Square DF Pr > ChiSq Likelihood Ratio 0.0000 5 1.0000 Score 269.3073 5 <.0001 Wald 0.0000 5 1.0000 Deviance Residuals for Mayo Model 15:44 Friday, May 9, 2008 4 The PHREG Procedure WARNING: The validity of the model fit is questionable. Analysis of Maximum Likelihood Estimates Parameter Standard Hazard Parameter DF Estimate Error Chi-Square Pr > ChiSq Ratio logbili 1 0 0.11375 0.0000 1.0000 1.000 logalb 1 0 0.92663 0.0000 1.0000 1.000 age 1 0 0.00912 0.0000 1.0000 1.000 logproth 1 0 1.17891 0.0000 1.0000 1.000 edema 1 0 0.51531 0.0000 1.0000 1.000 Martingale Residuals 15:44 Friday, May 9, 2008 5 Null Model vs. Transformed covars The PHREG Procedure Model Information Data Set WORK.A Dependent Variable time Censoring Variable delta Censoring Value(s) 0 Ties Handling BRESLOW Number of Observations Read 312 Number of Observations Used 312 Summary of the Number of Event and Censored Values Percent Total Event Censored Censored 312 125 187 59.94 Convergence Status Convergence criterion (GCONV=1E-8) satisfied. Model Fit Statistics Without With Criterion Covariates Covariates -2 LOG L 1279.960 1114.549 AIC 1279.960 1124.549 SBC 1279.960 1138.690 Testing Global Null Hypothesis: BETA=0 Test Chi-Square DF Pr > ChiSq Likelihood Ratio 165.4109 5 <.0001 Score 288.8856 5 <.0001 Wald 186.3233 5 <.0001 Martingale Residuals 15:44 Friday, May 9, 2008 6 Null Model vs. Transformed covars The PHREG Procedure Analysis of Maximum Likelihood Estimates Parameter Standard Hazard Parameter DF Estimate Error Chi-Square Pr > ChiSq Ratio Label bili 1 0.11634 0.01512 59.2310 <.0001 1.123 Serum Bilirubin, mg/dl logalb 1 -3.79843 0.70591 28.9537 <.0001 0.022 age 1 0.03287 0.00928 12.5547 0.0004 1.033 logproth 1 3.37936 0.90811 13.8481 0.0002 29.352 edema 1 0.79174 0.30918 6.5578 0.0104 2.207 Supremum Test for Functional Form Maximum Absolute Pr > Variable Value Replications Seed MaxAbsVal bili 26.1156 1000 19 <.0001 Martingale Residuals 15:44 Friday, May 9, 2008 7 Null Model vs. Transformed covars The PHREG Procedure Model Information Data Set WORK.A Dependent Variable time Censoring Variable delta Censoring Value(s) 0 Ties Handling BRESLOW Number of Observations Read 312 Number of Observations Used 312 Summary of the Number of Event and Censored Values Percent Total Event Censored Censored 312 125 187 59.94 Convergence Status Convergence criterion (GCONV=1E-8) satisfied. Model Fit Statistics Without With Criterion Covariates Covariates -2 LOG L 1279.960 1080.825 AIC 1279.960 1090.825 SBC 1279.960 1104.966 Testing Global Null Hypothesis: BETA=0 Test Chi-Square DF Pr > ChiSq Likelihood Ratio 199.1349 5 <.0001 Score 269.3073 5 <.0001 Wald 198.2020 5 <.0001 Martingale Residuals 15:44 Friday, May 9, 2008 8 Null Model vs. Transformed covars The PHREG Procedure Analysis of Maximum Likelihood Estimates Parameter Standard Hazard Parameter DF Estimate Error Chi-Square Pr > ChiSq Ratio logbili 1 0.87921 0.09873 79.2980 <.0001 2.409 logalb 1 -3.05332 0.72408 17.7818 <.0001 0.047 age 1 0.03327 0.00866 14.7565 0.0001 1.034 logproth 1 3.01567 1.02380 8.6764 0.0032 20.403 edema 1 0.78470 0.29913 6.8814 0.0087 2.192 Supremum Test for Functional Form Maximum Absolute Pr > Variable Value Replications Seed MaxAbsVal logbili 7.9703 1000 19 0.1950