Question in question: Is there any relation between intersection in set theory and correlation in statistics?

It is closed as lacking details or clarity. According to one mod, the question "might be answerable, but there are a lot of different possible answers, so it would help to rephrase that question into something more specific." However, I don't know what I should include more. Can you give some examples on how this question can be answered in different ways, so I can choose the one I'm interested most?


1 Answer 1


You should only ask practical, answerable questions based on actual problems that you face. Chatty, open-ended questions diminish the usefulness of our site and push other questions off the front page

(from our help on asking questions).

The set theory/correlation question strikes me as being of this type. I said as much in comments, but because the space for explanations is limited there, here I will elaborate a little.

Set theory is fundamental to mathematics, so it would not be unfair to draw some parallels between the subject question and hypothetical questions one can think of asking about how correlation (or some other statistical concept) might be "related" to, say, addition or logic or real numbers. These hypotheticals might help make the breadth and vagueness of such questions more apparent. In particular,

  • What form of "correlation" do you have in mind? Pearson, Spearman, Kendall's Tau, etc.? Some generic concept of "linear relation"? Or even a broad concept of "statistically related"?

  • What kind of "relation" are you considering and why might it be important? Many correlation coefficient formulas involve addition in some form, but in most cases it's doubtful this sheds any light either on the concept of correlation or of addition.

  • Addition can be exploited in many different ways, according to the many different formulas for the many different forms of correlation. Which way is of interest and why?

Another problem with the original question is that if such questions were considered on-topic here, we would open ourselves to moderating and curating an endless number of vague, open-ended questions of the form "is there any relation between [broad mathematical concept X] and [broad statistical concept] Y?" No such question is sufficiently specific and focused to be appropriate here on CV.

For more guidance, please read the remainder of the help page I cited.

  • 2
    $\begingroup$ I see. I'd thought that there is only one concept of correlation and one concept of intersection. It seems that I have to role up my sleeves to read more, because if I don't have the sufficient vocabulary then even experts can't see how to answer the question. That's sad because I expect experts can help me unclear it. $\endgroup$
    – Ooker
    May 10, 2021 at 17:21
  • 5
    $\begingroup$ I won't rise to that bait--you're coming awfully close to trolling us, now. $\endgroup$
    – whuber Mod
    May 10, 2021 at 17:22
  • 6
    $\begingroup$ Trolling? Can you tell me which part I'm trolling? I just sharing my emotion for what I experience. I don't blame you, or beg you to help me. I just state a phenomenon happening many times in many Stack Exchange sites when a novice asking a question and the experts see it's unclear. You may read another example in Biology. Sharing experience and emotion isn't trolling. $\endgroup$
    – Ooker
    May 10, 2021 at 17:33

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .