Proc Corr - One variable with all others

AD — Mon, 07 Jan 2013 20:53:37 GMT

I have a large dataset and I am interested in the pearson correlations between one variable (let's call it Var1) and all the other variables in the dataset (Var2, Var3,... , VarN). I am not interested in all the correlations between all the "other" variables (e.g., Var2 with Var3, Var2 with Var4, Var2 with VarN, etc...).

I realize that if I include all the variables in the Var statement in Proc Corr that I can get all the information I am looking for. And I can use ODS to output the results and data steps to get rid of the information I don't need. The reason for this post is that I don't SAS to have to compute all those correlations if possible, as this is a really large dataset and will require a lot of processing time and power.

Re: Proc Corr - One variable with all others

PGStats — Mon, 07 Jan 2013 21:15:16 GMT

Use the WITH statement :

proc corr data=myData;

var Var1;

with var2-var99;

run;

Re: Proc Corr - One variable with all others

AD — Tue, 08 Jan 2013 16:33:33 GMT

Thanks!

topic Re: Proc Corr - One variable with all others in SAS Programming

Proc Corr - One variable with all others

Re: Proc Corr - One variable with all others

Re: Proc Corr - One variable with all others