2
$\begingroup$

In my opinion, the question https://physics.stackexchange.com/questions/47215/where-can-i-learn-about-transforming-uniform-random-distribution-into-other-dis?noredirect=1 is on topic here. That exact question is asked and answered in courses titled "Computational Physics" when they discuss Monte Carlo methods. Anyone interested in that question is vastly more likely to be a physicist than a statistician. Finally, the question received two very answers from physicists. The probabilities involved in the problem are not particularly difficult.

To clarify, I think it is a physics question because it is about a commonly-used physics tool. As a comparison, I use a differential refractometer in my research. A question about the internal functioning of a refractometer would clearly be on-topic. I see questions about the internal functioning of Monte Carlo methods to be on the same level: a question about how a particular physics tool works. I agree, though, that it is on-topic for SciComp; if it's off-topic here, it would be better there than anywhere else.

$\endgroup$
1
  • 2
    $\begingroup$ It also received one very wrong answer in a comment from a physicist. It turns out the same question had been asked already and answered very nicely on stats.SE, so stats.SE is clearly a good home for it. Should we duplicate all math, stats, etc., questions that are relevant to physics on physics.SE? $\endgroup$ – Peter Shor Dec 20 '12 at 16:38
3
$\begingroup$

We have generally---though admittedly not always---directed questions about the mathematical tools of physics to Math.SE when the phrasing of the question is divorced from it's physical context. Now I see sampling as a computational tool for doing physics.

As such I would personally suggest SciComp.SE, Math.SE or Stats.SE (and note that the general answer has been given repeated on StackOverflow in the context of many specific questions) in that order.

$\endgroup$
2
$\begingroup$

What makes a question on topic here is not whether it is of interest to physicists, but whether it is about physics. I don't see any way in which that question is about physics.

For what it's worth, I would think that much (not all) of the content of a typical computational physics course, including Monte Carlo methods, would actually fall under the scope of Computational Science.

$\endgroup$
4
  • $\begingroup$ When I suggested Stats, I also thought about SciComp but figured it would get more traffic here than SciComp so recommending it move there wouldn't be an improvement really... $\endgroup$ – tpg2114♦ Dec 19 '12 at 20:16
  • 2
    $\begingroup$ Actually, to be honest, I would have considered sending it to Mathematics. But stats may not be the worst place for it. We'll see if they reject it. $\endgroup$ – David Z Dec 19 '12 at 20:17
  • $\begingroup$ I had all 3 in my comment draft and then scaled it back to just Stats :) $\endgroup$ – tpg2114♦ Dec 19 '12 at 20:18
  • $\begingroup$ It should better have gone to mathematics $\endgroup$ – Dilaton Dec 19 '12 at 22:12

You must log in to answer this question.

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