Logistic Regression: A Self-learning Text, Third Edition (Statistics in the Health Sciences)

The output produced by PROC LOGISTIC follows:

The LOGISTIC Procedure

Model Information

Data Set REF.EVANS

Response Variable chd

Number of Response Levels 2

Number of Observations 609

Link Function Logit

Optimization Technique Fisher's scoring

Response Profile

Ordered

Value CHD Count

1171

2 0 538

Model Fit Statistics

Criterion

Intercept

Only

Intercept and

Covariates

AIC 440.558 365.230

SC 444.970 404.936

2 Log L 438.558 347.230

Analysis of Maximum Likelihood Estimates

Parameter DF

Standard
Estimate Error Chi-Square Pr>ChiSq
Intercept 1 4.0497 1.2550 10.4125 0.0013
CAT 1 12.6894 3.1047 16.7055 <.0001
AGE 1 0.0350 0.0161 4.6936 0.0303
CHL 1 0.00545 0.00418 1.7000 0.1923
ECG 1 0.3671 0.3278 1.2543 0.2627
SMK 1 0.7732 0.3273 5.5821 0.0181
HPT 1 1.0466 0.3316 9.9605 0.0016
CH 1 2.3318 0.7427 9.8579 0.0017
CC 1 0.0692 0.0144 23.2020 <.0001

Odds Ratio Estimates

Effect

Point

Estimate

95% Wald

Confidence Limits

CAT <0.001 <0.001 0.001

AGE 1.036 1.003 1.069

CHL 0.995 0.986 1.003

ECG 1.444 0.759 2.745

SMK 2.167 1.141 4.115

HPT 2.848 1.487 5.456

CH 0.097 0.023 0.416

CC 1.072 1.042 1.102

604 Appendix: Computer Programs for Logistic Regression