Dirac delta is pretty basic concept and we have multiple questions mentioning it. It also causes many confusion, so it is likely to provoke new questions among those who first encounter it. Maybe we should create a tag for it?

  • 2
    $\begingroup$ Delta function is a standard mathematical object and corresponding tag exists on Math.SE (with 500+ questions). But does it have any particular relevance for statistics to warrant a tag here? $\endgroup$
    – amoeba
    Oct 8, 2016 at 19:35
  • 3
    $\begingroup$ @amoeba it is used in statistics and probability and seems to cause some confusion for those who first encountered it. Moreover, in probability it can be regarded as a degenerate distribution and understood as it it provokes on-topic questions. $\endgroup$
    – Tim
    Oct 9, 2016 at 8:26
  • 1
    $\begingroup$ The Dirac delta is implicit to many stochastic processes as it is the identity function for convolution. However, it would not be used alone, but always in conjunction with something else. I do not personally see any particular reason for it to appear as a stand alone tag. $\endgroup$
    – Carl
    Oct 14, 2016 at 23:33

1 Answer 1


I am providing an answer so to bump the question and close it with some conclusion.

Personally I'd considered [dirac-delta] tag useful. The question already got eight upvotes and one negative (?) comment by Carl. I'd say that the response to this idea is positive, so I'd consider creating it.

So to close this issue: is there anyone who does not agree with this idea? Are there any more arguments against it?


Since there is not much interest in the thread and it got rather positive response, I created the tag and started re-tagging.

  • 2
    $\begingroup$ I don't necessarily have a problem w/ the growth of the number of tags, or having tags that are only applied to a few threads. I think tags should be intrinsically meaningful, have excerpts, & not be duplicates / synonyms. If you think it would be useful, you should make the tag. $\endgroup$ Oct 25, 2016 at 19:00
  • $\begingroup$ So, we promote the identity function for convolution to tag status, but not the identity meta.stats.stackexchange.com/questions/4440/…. So, we name individual leaves, and ignore their tree. What, I muse, would we do with a whole forest? LOL. $\endgroup$
    – Carl
    Nov 19, 2016 at 6:38

You must log in to answer this question.

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