Cross Tabulation of Two variables

[jeka1212]

Hi all,

I need  to create a combined binary variable beta1_beta2  with y1 and y2 and use the existing binary variable “kyphosis”  with proc beta1_beta2  *kyphosis.  I was wondering if you can advise me how to modify the SAS code below to repeat the process for all pair (y1,y2) in the dataset? 

Thanks,

data have;
infile "&sasforum\datasets\wiedat2b.csv" firstobs=2 dsd;
input y1 y2 kyphosis;
run;

DATA have;
set have;

********Creating the first binary variable beta1 with y1=28;
if y1 <= 28 then beta1=0;
else beta1=1; 

**Creating the first binary variable beta2 with y2=13.3;
if y2 <=13.3 then beta2=0;
else beta2=1;

********Creating the first  combined binary variable beta1_beta2 with the pair (y1=28,y2=13.3);

if beta1=1 & beta2=1 then beta1_beta2=1;
if beta1=1 & beta2=0 then beta1_beta2=1;
if beta1=0 & beta2=1 then beta1_beta2=1;
if beta1=0 & beta2=0 then beta1_beta2=0;
run;

proc freq data = have order=data noprint ;
            tables  beta1_beta2*kyphosis / out=FreqCount OUTPCT sparse ;
		*output out=mnocol;
run;

*THEN REPEAT  FOR:  y1=28, y2=11.1, 12.6, ect...; 

***Do the same for the second paire y1=28,y2=11.1;
***Do the same for the second paire y1=28,y2=12.6;
**........;
***Do the same for the last (141) paire y1=28,y2=14.2;


**THEN REPEAT FOR: y1=16.7, y2=11.1, 12.6, ect...; 

***Do the same for the second paire y1=16.7,y2=11.1;
***Do the same for the second paire y1=16.7,y2=12.6;
**........;
***Do the same for the last (141) paire y1=16.7,y2=14.2;

****HOW TO REPEAT THIS FROCESS FOR ALL 19881 COMBINATION IN THE DATASET (141 X 141 POSSIBLE COMBINATIONS)

[ChrisNZ]

Liek this?

data HAVE;
  input Y1 Y2 KYPHOSIS;
cards;
28 11.1 100
28 13.3 200
16 16.1 300
16 13.3 400
run;

data _null_;
  set HAVE(keep=Y1) nobs=NOBS;
  do I=1 to NOBS;
    J+1;
    set HAVE(keep=Y2) point=I;
    call execute('data TEMP; set HAVE;');
    call execute(catt('BETA1_BETA2=(Y1<=',Y1,')|(Y2<=',Y2,');'));
    call execute('proc freq data=TEMP order=data noprint;');
    call execute('tables BETA1_BETA2*KYPHOSIS');
    call execute(catt('/out=OUTFREQ',J,'(label="Y1=',Y1,' Y2=',Y2,'") outpct sparse;run;'));
  end;
run;



 

[jeka1212 · Accepted solution]

Many thanks ChrisNZ. The codes did help. 

I was wondering if you can help to append  each results in 1 big dataset please?  I have attached an example.

data HAVE;
  input Y1 Y2 KYPHOSIS;
cards;
28 11.1 100
28 13.3 200
16 16.1 300
16 13.3 400
run;

data _null_;
  set HAVE(keep=Y1) nobs=NOBS;
  do I=1 to NOBS;
    J+1;
    set HAVE(keep=Y2) point=I;
    call execute('data TEMP; set HAVE;');
    call execute(catt('BETA1_BETA2=(Y1<=',Y1,')|(Y2<=',Y2,');'));
    call execute('proc freq data=TEMP order=data noprint;');
    call execute('tables BETA1_BETA2*KYPHOSIS');
    call execute(catt('/out=OUTFREQ',J,'(label="Y1=',Y1,' Y2=',Y2,'") outpct sparse;run;'));
  end;
run;

[ChrisNZ]

Microsoft documents are best avoided.

Many people will not touch them.

Just add this:

    call execute(catt('proc append base=BASE append=OUTFREQ',J,';run;'));

[jeka1212]

OK many thanks.

I am having the error message below:

NOTE: Line generated by the CALL EXECUTE routine.
96 + proc append base=BASE append=OUTFREQ16;run;
------
22
76

ERROR 22-322: Syntax error, expecting one of the following: ;, (, APPENDVER, APPENDVERSION,
BASE, CREATE, DATA, FORCE, GETSORT, NEW, NOWARN, OUT.

ERROR 76-322: Syntax error, statement will be ignored.

Basically each paire of (y1,y1) has "OUTFREQ" output like this:

Beta1_beta2	kyphosis	COUNT	PERCENT	PCT_ROW	PCT_COL
0	1	1	0.70922	100	1.111111
0	0	0	0	0	0
1	1	89	63.12057	63.57143	98.88889
1	0	51	36.17021	36.42857	100

I need to append it like this: 

Y1	Y2	perctCount POS	perct count NEG	FALPOS	FALNEG	SENSIT	SPECIF	NPV	PPV
16	13.3	89	0	1	51	98.89	0	0	63.57143

Any help? 

[ChrisNZ]

>proc append base=BASE append=OUTFREQ16;run;

My bad, end of Friday evening error. data= not append=.

Have even bothered looking at the syntax for proc append? You have to help yourself too.

As for reformatting the dataset, you can add an extra step after table BASE is created.

[jeka1212]

I corrected the error after I posted the question. Thanks again

Still working on reformatting now

[ChrisNZ]

> I corrected the error after I posted the question. 

You should have edited your question then, to avoid my spending time giving you an answer you already had.

[jeka1212]

Sorry

[ChrisNZ]

No worries. For next time. 🙂

[jeka1212]

is your book (High-Performance SAS Coding) good for beginners in programming  like us?

[ChrisNZ]

This book is not targeted at beginners. The SAS language itself is not taught. But as you learn it, the book shows how to use the language's features efficiently and explains what happens behind the scene. There are many ways to perform a given task with SAS. Some are much more efficient than others, 

[jeka1212]

OK good to know. Will order one as soon as possible

[ChrisNZ]

Thank you. I hope you find it useful.

[ballardw]

beta1 = y1>28;
beta2 = y2>13.3;

is another way to assign the beta variables.

Do you really want to assign 0 if y1 or y2 are actually missing though?

if beta1=1 & beta2=1 then beta1_beta2=1;
if beta1=1 & beta2=0 then beta1_beta2=1;
if beta1=0 & beta2=1 then beta1_beta2=1;
if beta1=0 & beta2=0 then beta1_beta2=0;

(though else would help on the 2nd 3rd and 4th If)

is the same as

beta1_beta2= beta1 or beta2;

which will execute a tad faster.

Which will matter on doing 19881 iterations.

You better have something that saves your Proc Freq output to a unique name each time or each iteration will overwrite the previous