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

Criteria for Assessing Goodness of Fit

Criterion DF Value Value/DF

Deviance 600 347.2295 0.5787

Scaled Deviance 600 347.2295 0.5787

Pearson Chi-Square 600 799.0652 1.3318

Scaled Pearson X2 600 799.0652 1.3318

Log Likelihood 173.6148

Algorithm converged

Analysis of Parameter Estimates

Parameter Estimate

Standard

Error

Wald 95%

Confidence Limits

Chi-

Square Pr>ChiSq

Intercept 4.0497 1.2550 6.5095 1.5900 10.41 0.0013
CAT 12.6895 3.1047 18.7746 6.6045 16.71 <.0001
AGE 0.0350 0.0161 0.0033 0.0666 4.69 0.0303
CHL 0.0055 0.0042 0.0137 0.0027 1.70 0.1923
ECG 0.3671 0.3278 0.2754 1.0096 1.25 0.2627
SMK 0.7732 0.3273 0.1318 1.4146 5.58 0.0181
HPT 1.0466 0.3316 0.3967 1.6966 9.96 0.0016
CH 2.3318 0.7427 3.7874 0.8762 9.86 0.0017
CC 0.0692 0.0144 0.0410 0.0973 23.20 <.0001
Scale 1.0000 0.0000 1.0000 1.0000

NOTE: The scale parameter was held fixed.

Contrast Estimate Results

Label

L'Beta

Estimate

Standard

Error Confidence Limits

Chi-

Square

Pr>

ChiSq

Log OR (ch1¼220,
hpt¼1)

0.1960 0.4774 0.7397 1.1318 0.17 0.6814

Exp(LogOR(chl¼220,
hpt¼1))

1.2166 0.5808 0.4772 3.1012

Log OR (chl¼220,
hpt¼0)

2.5278 0.6286 1.2957 3.7599 16.17 <.0001

Exp(LogOR(chl¼220,
hpt¼0))

12.5262 7.8743 3.6537 42.9445

Contrast Results

Contrast DF Chi-Square Pr>ChiSq Type

LRT for interaction terms 2 53.16 <.0001 LR

The table titled “Contrast Estimate Results” gives the odds ratios requested by the
ESTIMATE statement. The estimated odds ratio for CAT¼1 vs. CAT¼0 for a

SAS 607