Use the LOGISTIC macro.

1. Issue the following commands to SAS (with appropriate
   substitutions for "U:\").

LIBNAME GFLib 'U:\dm\sasdata' ; 

LIBNAME RCLib 'U:\dm\RC' ;

2. If necessary, apply macros to convert Excel data to
SAS (EXCELSAS, see MacroInstr.txt), to divide the data 
into training/test or training/validation/test subsets
(RANSPLIT, see MacroInstr.txt), or to explore the data
(UNIVAR and FREQ/FREQUENCY, see MacroInstr2.txt).

3. If necessary, assign numerical designations to 
categories and shorten variable names so that you can 
fit them all in the appropriate fields of the LOGISTIC
macro.  For example:

DATA RCLIB.SATRAIN2;
SET RCLIB.SATRAIN;
IF FAMHIST = 'Present' THEN FAM=1;
IF FAMHIST = 'Absent' THEN FAM=0;
to = tobacco;
ad = adiposity;
al = alcohol;
ta = typea;
ob = obesity;
RUN;

DATA RCLIB.SAVALID2;
SET RCLIB.SAVALID;
IF FAMHIST = 'Present' THEN FAM=1;
IF FAMHIST = 'Absent' THEN FAM=0;
to = tobacco;
ad = adiposity;
al = alcohol;
ta = typea;
ob = obesity;
RUN;

DATA RCLIB.SATEST2;
SET RCLIB.SATEST;
IF FAMHIST = 'Present' THEN FAM=1;
IF FAMHIST = 'Absent' THEN FAM=0;
to = tobacco;
ad = adiposity;
al = alcohol;
ta = typea;
ob = obesity;
RUN;

4. Apply the LOGISTIC macro to your training data set.
How to fill in the blanks is described on pages 187-191
of the textbook.  I recommend converting non-dichotomous
categorical variables to dichotomous indicators and then
including both the continuous variables and dichotomous
indicators in the fourth and fifth fields.  (Filling in
the third field changes the kind of output you get.)
 
	For output like that in {RCLIB.SATRAIN224.rtf}
	(i.e., to obtain a fitted model and suggestions
	about which variables can be removed), try:
	
	RCLIB.SATRAIN2		chd
	[blank]			sbp to ldl ad fam ta ob al age
	sbp to ldl ad fam ta ob al age
	[blank]			none
	[blank]			[blank]
	subject			24
	word			U:\dm\RC\
	[blank]			0.50
			
	For output like that in {RCLIB.SATRAIN222.rtf}
	(i.e., to obtain diagnostic plots suggesting
 	which quadratic/interaction terms may help), try:

	RCLIB.SATRAIN2		chd
	[blank]			sbp to ldl ad fam ta ob al age
	sbp to ldl ad fam ta ob al age
	yes			none
	[blank]			[blank]
	subject			22
	word			U:\dm\RC\
	[blank]			0.50

 


The LSCORE macro does not work!  You can use the code
below to obtain correct classification rates for the 
validation (or test) subset.  If you insert the line
	WHERE CHD = 1;
immediately after the line starting with  PROC MEANS,
the code will give you sensitivities instead of 
correct classification rates.  If you insert the line
	WHERE CHD = 0;
immediately after the line starting with  PROC MEANS,
the code will give you specificities instead of
correct classification rates.  

For the validation subset in {savalid2.sas7bdat}, the
correct classification rates range from 35.7% (when
we predict CHD for any subject with estimated risk
at least 5%) to 73.9% (when we predict CHD for any
subject with estimated risk at least 50%).  The
sensitivities decline from 97.4% (when we predict
CHD for any subject with estimated risk at least 5%)
to 0% (when we predict CHD for any subject with 
estimated risk at least 95%).  

DATA RCLIB.SAVALID2;
	* Change RCLIB.SAVALID2 to the name of your validation (or test) subset. ;
SET  RCLIB.SAVALID2;
ESTLOGITRISK = -6.9699 + SBP*0.0169 + TO*0.1255 + LDL*0.1399 + AD*0.0479 + FAM*0.9254 + TA*0.0383 + OB*-0.0584 + AL*-0.00531 + AGE*0.0163;
	* This fitted model is shown on page 15 of {RCLIB.SATRAIN224.rtf}. ;
ESTRISK = exp(ESTLOGITRISK)/(1+exp(ESTLOGITRISK));
PREDWITHCUTOFF05 = 1 - (ESTRISK < 0.05);
PREDWITHCUTOFF10 = 1 - (ESTRISK < 0.10);
PREDWITHCUTOFF15 = 1 - (ESTRISK < 0.15);
PREDWITHCUTOFF20 = 1 - (ESTRISK < 0.20);
PREDWITHCUTOFF25 = 1 - (ESTRISK < 0.25);
PREDWITHCUTOFF30 = 1 - (ESTRISK < 0.30);
PREDWITHCUTOFF35 = 1 - (ESTRISK < 0.35);
PREDWITHCUTOFF40 = 1 - (ESTRISK < 0.40);
PREDWITHCUTOFF45 = 1 - (ESTRISK < 0.45);
PREDWITHCUTOFF50 = 1 - (ESTRISK < 0.50);
PREDWITHCUTOFF55 = 1 - (ESTRISK < 0.55);
PREDWITHCUTOFF60 = 1 - (ESTRISK < 0.60);
PREDWITHCUTOFF65 = 1 - (ESTRISK < 0.65);
PREDWITHCUTOFF70 = 1 - (ESTRISK < 0.70);
PREDWITHCUTOFF75 = 1 - (ESTRISK < 0.75);
PREDWITHCUTOFF80 = 1 - (ESTRISK < 0.80);
PREDWITHCUTOFF85 = 1 - (ESTRISK < 0.85);
PREDWITHCUTOFF90 = 1 - (ESTRISK < 0.90);
PREDWITHCUTOFF95 = 1 - (ESTRISK < 0.95); 
CORRECT05 = (PREDWITHCUTOFF05 = CHD);
	* Change CHD to the name of your response variable. ;
CORRECT10 = (PREDWITHCUTOFF10 = CHD);
CORRECT15 = (PREDWITHCUTOFF15 = CHD);
CORRECT20 = (PREDWITHCUTOFF20 = CHD);
CORRECT25 = (PREDWITHCUTOFF25 = CHD);
CORRECT30 = (PREDWITHCUTOFF30 = CHD);
CORRECT35 = (PREDWITHCUTOFF35 = CHD);
CORRECT40 = (PREDWITHCUTOFF40 = CHD);
CORRECT45 = (PREDWITHCUTOFF45 = CHD);
CORRECT50 = (PREDWITHCUTOFF50 = CHD);
CORRECT55 = (PREDWITHCUTOFF55 = CHD);
CORRECT60 = (PREDWITHCUTOFF60 = CHD);
CORRECT65 = (PREDWITHCUTOFF65 = CHD);
CORRECT70 = (PREDWITHCUTOFF70 = CHD);
CORRECT75 = (PREDWITHCUTOFF75 = CHD);
CORRECT80 = (PREDWITHCUTOFF80 = CHD);
CORRECT85 = (PREDWITHCUTOFF85 = CHD);
CORRECT90 = (PREDWITHCUTOFF90 = CHD);
CORRECT95 = (PREDWITHCUTOFF95 = CHD);
RUN;

TITLE ' ';
PROC MEANS DATA=RCLIB.SAVALID2 MEAN;
VAR CORRECT05 CORRECT10 CORRECT15 CORRECT20 CORRECT25 CORRECT30 CORRECT35 CORRECT40 CORRECT45 CORRECT50 CORRECT55 CORRECT60 CORRECT65 CORRECT70 CORRECT75 CORRECT80 CORRECT85 CORRECT90 CORRECT95;
RUN;