Multiple Frequency Table

[Jamie_H]

I have a table that contains individual rows of data for customers, with various other variables that have a 1 or a 0.  You can use this to get a feel for what I mean:

data have;
input ID ColumnA StatusB CategoryF UnderPerf;
datalines ;
1 1 0 0 1
2 1 0 1 1
3 0 1 1 1
4 0 0 0 0
5 1 1 1 1
;
run;

What I would like off the back of this is to get a frequency of each of the variables by each other in one big matrix such as this:

Want

Please bear in mind that these are made up variables, and in total there about 20. 

Many thanks!

Jamie

[PaigeMiller]

I do not grasp the logic that lets you go from input data to output table. Please explain.

[ed_sas_member · Accepted solution]

HI @Jamie_H 

You can use PROC CORR to achieve this (cf. similar post https://communities.sas.com/t5/SAS-Programming/Calculating-Product-Overlap-in-a-Basket/m-p/631529#M1... and great @Ksharp solution!)

proc corr data=have out=have_corr sscp noprint ;
	var ColumnA StatusB CategoryF UnderPerf;
run;

data want;
	set have_corr;
	where _type_='SSCP' and _name_ ne 'Intercept';
	keep _name_ ColumnA StatusB CategoryF UnderPerf;
run;

Capture d’écran 2020-03-16 à 16.40.01.png

[Jamie_H]

That's perfect Ed and so useful to get to play with a Proc that I had heard of before but never had the opportunity to use. 

Many thanks!