13
$\begingroup$

I am reading a paper on multiple change-point model and do not understand one of the equation.

Is it considered on-topic to ask questions about a paper like that? On the one hand, it's clearly statistics. On the other hand, the question may be so specific that it may not contribute towards "building a knowledge site". Plus, I'm not sure about the community's reception towards question about specific paper (vs. question about general statistical concept.)

$\endgroup$
21
$\begingroup$

I don't see any necessary problem with asking a question like that. Here are some things to be sure you do:

  1. Provide a complete and correct citation for the paper.
  2. Provide a link to the paper.
  3. Paste into the body of your question the equation / text you need help understanding (i.e., don't just say 'I need help with equation 5').
  4. Paste in enough of the context that the question can stand on its own and potential answerers don't need to read the paper to answer your question, if they don't want to.

I recognize that the existence of 4 seems inconsistent with 1 and 2, but you should do all four.

N.B., you say that your question is statistical. I have no reason not to take you at your word. But if your question were only about, say, the derivation of the equation, 'how does equation $i$ fall out of equation $i-1$', it may be a better fit on SE's math site. Likewise, if it were about the economic implications of the equation, it may be a better fit on SE's economics site, etc.

$\endgroup$
10
$\begingroup$

I think it's fine. After all, a "knowledge site" doesn't mean a statistical textbook or encyclopaedia—those exist already, & answering specific, even idiosyncratic, questions is almost the raison d'être of CV, certainly its forte. If you follow @gung's advice the question will be as potentially useful to other people as it can be; if it happens not to help anyone else, no harm done.

$\endgroup$

You must log in to answer this question.

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