Correlation analysis with multiple variables

[Babloo]

Could you please help me understand how I can do the Correlation analysis with 200 variables?

[PGStats]

Any particular problem with proc corr documentation?

[Babloo]

no, can we do proc corr with 200 variables? Is it worth doing that?

[PGStats]

Sure you can. Whether you should depends on your objective and on the nature of the data. Be aware that results can be difficult to interpret when your data has many missing values.

[PaigeMiller]

---

@Babloo wrote:

no, can we do proc corr with 200 variables? Is it worth doing that?

---

The real problem is not whether or not PROC CORR can do this. Yes, of course it can. The real problem is: what use of this 200x200 correlation matrix do you have in mind? It's hard to see how to make use of a 200x200 correlation matrix for additional analyses, unless you have something automated in mind (what?)

[Reeza]

---

@Babloo wrote:

no, can we do proc corr with 200 variables? Is it worth doing that?

---

Try it.