/* Example of Logistic Regression */
OPTIONS LINESIZE=70;
PROC FORMAT;
  VALUE agegroup 0 = '<= 20 and <= 65'
                 1 = '<  20 or  >  65';
  VALUE vision   0 = 'No Problem'
                 1 = 'Some Problem';
  VALUE yes_no   0 = 'No'
                 1 = 'Yes';
RUN;

DATA logistic;
  INFILE 'accident dat';
  INPUT accident 1. age 2. vision 1. drive_ed 1.;
  IF age < 20 OR age > 65 THEN agegroup = 1;
  ELSE agegroup = 0;
  IF age < 20 THEN young = 1;
  ELSE young = 0;
  IF age > 65 THEN old = 1;
  ELSE old = 0;

LABEL accident = 'Accident in Last Year?'
      age      = 'Age of Driver'
      vision   = 'Vision Problem?'
      drive_ed = 'Driver Education?'

FORMAT accident
       drive_ed
       young
       old      yes_no.
       agegroup agegroup.
       vision   vision.;
RUN;

PROC LOGISTIC DATA=LOGISTIC DESCENDING;
  TITLE 'Predicting Accidents Using Logistic Regression';
  MODEL accident = age vision drive_ed /
                   SELECTION = FORWARD
                   CTABLE PPROB= (0 TO 1 BY .1)
                   LACKFIT
                   RISKLIMITS;
  OUTPUT OUT=pred P=phat LOWER=lcl UPPER=ucl;
RUN;
PROC SORT DATA=pred OUT=sortpred;
  BY phat;
RUN;
PROC PRINT DATA=sortpred;
RUN;