8
$\begingroup$

It's been a long time since I've been active on CV and I don't want to botch it by asking an off-topic, opinion-based, or duplicate question. Here is my attempt:

How to answer to critiques about the inapplicability of the framework of frequentist statistics to the real world?

I often hear the argument that frequentist stats is useless or contorted because no event is precisely repeatable, let alone repeatable infinitely many times, and because there are no iid sequences in the real world, since "causal factors cause distributions shifts (?!)". Here is an example of the egregious use of this sort of arguments:

http://mitchgordon.me/math/2021/04/02/probability.html

Having applied frequentist stat with success to real world problems for years, I couldn't disagree more with the argument(s). Can you help me finding the weak points of these arguments?

For the record, I used to learn a few good answers against these arguments when I first studied probability theory, but I don't remember them at the moment.

$\endgroup$
6
  • 5
    $\begingroup$ At first blush, this seems reasonable to me. It might be worth quoting the explicit arguments where you want help finding the weak points. $\endgroup$ – gung - Reinstate Monica Apr 3 at 12:51
  • $\begingroup$ Thanks, I'll proceed to ask then. $\endgroup$ – DeltaIV Apr 4 at 16:56
  • $\begingroup$ Done: stats.stackexchange.com/questions/518183/… $\endgroup$ – DeltaIV Apr 4 at 17:51
  • $\begingroup$ Now that the question is posted on the main site. and fine, does this Meta question have enduring value? I know there's an accepted answer, and thanks to @gung for saying what I guess many of us would like to have said, but it now seems redundant. $\endgroup$ – Nick Cox Apr 7 at 17:10
  • $\begingroup$ @NickCox, at this point, the mods would have to delete this thread. You can flag it, if you want. (FWIW, I'm OK w/ it, if they say 'yes'.) $\endgroup$ – gung - Reinstate Monica Apr 7 at 17:29
  • $\begingroup$ @gung-ReinstateMonica Thanks for that, but as it's not a nuisance thread and everyone who has contributed is in very good standing I'd rather ask here first. I don't have any compunction in flagging threads that seem perverse or obnoxious, the complete opposite from what is here. $\endgroup$ – Nick Cox Apr 7 at 17:46
4
$\begingroup$

(Turning my comment into an answer.)

At first blush, this seems reasonable to me. It might be worth quoting the explicit arguments where you want help finding the weak points.

$\endgroup$
0
$\begingroup$

When you ask how to respond you already close the option for the answer to say they disagree with the notion and in fact think frequentism is intrinsically flawed. It doesn't get to the point that it's opinion based, but I find it's easier to ask: "Is frequentism flawed?" or "Why is frequentism valid?" - answers naturally come up with arguments to why frequentism is or isn't invalid.

$\endgroup$

You must log in to answer this question.

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