12
$\begingroup$

This question is put on hold as opinion based. The tldr version of the question is "Why are some people saying Frequentist statistics is no good?".

I have partial sympathy with putting it on hold. One the one hand, I don't actually think it's an opinion based question. The OP is not asking "Are Frequentist statistics no good?", but rather wants to really understand the reasoning behind that argument. I do admit that distinction is a grey line: someone might think a reasonable answer is "because they don't understand math", which would help no one (also not my opinion). Also, I do have the fear that it will could turn into a highly-opinion based set of answers.

With that said, one of the reasons I chose to answer it is that I think it's a very important topic worth discussing. I know there have been several somewhat similar questions (i.e. Bayesian vs Frequentist: pros and cons), but I do think this one is a bit more of a niche.

Is CV the right place for such a question? Or is it better suited for something like a blog post?

$\endgroup$
4
  • 3
    $\begingroup$ I see that I was the last person of 5 to vote against that question. I would have (could have) downvoted it on several grounds, not just that reported as the expressed community view. It's not that it doesn't touch on interesting and important issues: rather, its wording and its links more or less doomed it to be seem too broad as well as too contentious. $\endgroup$
    – Nick Cox
    Apr 4, 2018 at 17:27
  • $\begingroup$ I should add that I liked your answer and upvoted it. Contradictory? I guess so. $\endgroup$
    – Nick Cox
    Apr 4, 2018 at 18:19
  • $\begingroup$ @NickCox: good answer, wrong forum is a completely defendable position $\endgroup$
    – Cliff AB
    Apr 4, 2018 at 19:05
  • $\begingroup$ At this time, it's a tie between closing for reason A or closing for reason B. Seems like the correct course of action is pretty clear. $\endgroup$
    – Cliff AB
    Apr 5, 2018 at 15:35

3 Answers 3

10
$\begingroup$

I think it is debatable whether this Q is "primarily opinion-based" (our closing reason) or not, but it seems to me that it is clearly "too broad". It asks about classical statistics vs machine learning, then about frequentist statistics vs Bayesian statistics, and then about book recommendations on "good" statistics. A poor question that should be closed.

If I had had a chance to vote, I would have voted to close as too broad.

$\endgroup$
6
$\begingroup$

Yes, this is too opinion based for CV. A blog post would be more reasonable.

$\endgroup$
2
  • 4
    $\begingroup$ I'm mixed on this. It is (another) opportunity for petty B vs A pontificating, & we have several of those already, so it could be a duplicate. OTOH, your answer is balanced (& good, +1). However, it was voted to close by 5 CV users, including both Bayesians and frequentists, with high reputations & who contribute frequently. That's the standard the SE system uses. As a result, I'm inclined to let the closure stand. $\endgroup$ Apr 4, 2018 at 13:05
  • $\begingroup$ I agree that it is opinion based because what could possibly be a "right" answer here? The blogs that were linked speak for themselves. Would a "right" answer merely parrot the views expressed on the blogs? If the OP has trouble understanding the points in the blogs he/she linked, they should post asking for clarification, but that is not the question that was asked. $\endgroup$
    – AdamO
    Apr 5, 2018 at 15:06
-2
$\begingroup$

No, this is an important discussion topic. It belongs on CV.

$\endgroup$
3
  • 3
    $\begingroup$ Note: it would be better for people to only upvote their choice, & not also downvote the option they disagree with. That policy will yield the clearest picture of the community's preferences. $\endgroup$ Apr 4, 2018 at 15:17
  • 1
    $\begingroup$ @gung As we won't know who voted once, twice or three times interpretation is fraught any way.... $\endgroup$
    – Nick Cox
    Apr 4, 2018 at 17:23
  • 3
    $\begingroup$ @NickCox, it's a bit late now, but if no one had downvoted, the picture would be clearer. That's my point. $\endgroup$ Apr 4, 2018 at 17:41

You must log in to answer this question.

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