A CASE-CONTROL STUDY with R=3 CONTROLS PER CASE 1 Each row has one case (Case=1) followed by 3 controls (Case=0) COVARIATE AVERAGES BY CASE STATUS 18:37 Wednesday, December 7, 2005 The MEANS Procedure N case Obs Variable Mean Std Dev --------------------------------------------------------------- 0 150 gender 0.4466667 0.4988129 wunk 79.3666667 5.3362338 hist 0.3800000 0.4870125 1 50 gender 0.6000000 0.4948717 wunk 78.9000000 4.7175314 hist 0.6600000 0.4785181 --------------------------------------------------------------- A CASE-CONTROL STUDY with R=3 CONTROLS PER CASE 2 Each row has one case (Case=1) followed by 3 controls (Case=0) ANALYSIS BY PROC PHREG 18:37 Wednesday, December 7, 2005 `OBSERVED DEATHS' AT CASES, STRATIFIED BY SUBJECT The PHREG Procedure Model Information Data Set WORK.CASES Dependent Variable Time Censoring Variable case Censoring Value(s) 0 Ties Handling BRESLOW Number of Observations Read 200 Number of Observations Used 200 Convergence Status Convergence criterion (GCONV=1E-8) satisfied. Model Fit Statistics Without With Criterion Covariates Covariates -2 LOG L 138.629 121.380 AIC 138.629 127.380 SBC 138.629 133.116 Testing Global Null Hypothesis: BETA=0 Test Chi-Square DF Pr > ChiSq Likelihood Ratio 17.2494 3 0.0006 Score 16.4497 3 0.0009 Wald 14.5837 3 0.0022 Analysis of Maximum Likelihood Estimates Parameter Standard Hazard Variable DF Estimate Error Chi-Square Pr > ChiSq Ratio gender 1 0.61237 0.33785 3.2854 0.0699 1.845 wunk 1 -0.03546 0.03714 0.9116 0.3397 0.965 hist 1 1.34069 0.38757 11.9661 0.0005 3.822 A CASE-CONTROL STUDY with R=3 CONTROLS PER CASE 3 Each row has one case (Case=1) followed by 3 controls (Case=0) ANALYSIS BY PROC PHREG 18:37 Wednesday, December 7, 2005 GENDER AND HISTORY ONLY The PHREG Procedure Model Information Data Set WORK.CASES Dependent Variable Time Censoring Variable case Censoring Value(s) 0 Ties Handling BRESLOW Number of Observations Read 200 Number of Observations Used 200 Convergence Status Convergence criterion (GCONV=1E-8) satisfied. Model Fit Statistics Without With Criterion Covariates Covariates -2 LOG L 138.629 122.297 AIC 138.629 126.297 SBC 138.629 130.121 Testing Global Null Hypothesis: BETA=0 Test Chi-Square DF Pr > ChiSq Likelihood Ratio 16.3326 2 0.0003 Score 15.7428 2 0.0004 Wald 14.0183 2 0.0009 Analysis of Maximum Likelihood Estimates Parameter Standard Hazard Variable DF Estimate Error Chi-Square Pr > ChiSq Ratio gender 1 0.58691 0.33669 3.0386 0.0813 1.798 hist 1 1.31009 0.38436 11.6181 0.0007 3.707