-2
$\begingroup$

At the moment I'm reading some texts on Markov Chains, to learn the subject, and I'm getting the feeling that Math.SE community is not interested/very knowledgeable in this area, when compared to other more theoretical areas. Why is that? Is there a way to solve this?

I do not intend to disrespect anyone, with this question.

Cheers.

Edit: Just inputting some info. The unanswered-to-active ratio for some tags Algebraic Geometry (32%), Differential Geometry (29%), however, Stochastic Processes (37%).

$\endgroup$
6
  • $\begingroup$ Wouldn't these topics fit better on Cross Validated? $\endgroup$
    – Asaf Karagila Mod
    Commented Dec 12, 2015 at 17:22
  • 4
    $\begingroup$ @AsafKaragila Only when they involve making a statistical inference about an unknown distribution. The majority of stochastic processes questions operate with a known distribution, which places them within probability rather than statistics. Question count in stochastic-processes tag is 4478 on Math vs 417 on Stats. $\endgroup$
    – user147263
    Commented Dec 12, 2015 at 18:28
  • $\begingroup$ @Norma: I'll take your word for it. Statistics and probability were never my strong side. $\endgroup$
    – Asaf Karagila Mod
    Commented Dec 12, 2015 at 18:30
  • $\begingroup$ Just curious, how do you calculate the unanswered to active ratio? $\endgroup$
    – Asaf Karagila Mod
    Commented Dec 14, 2015 at 5:06
  • $\begingroup$ @AsafKaragila unanswered/active =P $\endgroup$ Commented Dec 14, 2015 at 14:31
  • 3
    $\begingroup$ Ah. Actually if you go to "top users" of the tag you also see a percentage of unanswered. I'm happy to report that in the tags I roam this percentage is particularly low. :-) $\endgroup$
    – Asaf Karagila Mod
    Commented Dec 14, 2015 at 14:45

2 Answers 2

4
$\begingroup$

I'm interested in stochastic processes and I rarely answer questions about stochastic processes on here

Most questions are (in my experience) either

a) very applied/computational.

b) tedious details about all the measure theoretic definitions of Brownian motion/Ito integral etc.

I don't want to do computational/applied stuff. I don't often like discussing all the tedious measure theory, as it is often a huge detailed mess.

Every now any then there will be a question about how to solve an SDE which is somewhat a happy medium but honestly it's not super interesting to me and can require playing around with Ito's lemma.

I think all the interesting questions from my perspective would be on mathoverflow and then I'm not quite developed enough to answer those too much. :)

I feel like it's because there's such a huge divide between computation and theory, and all the introductory theory is really tedious and often requires a much lengthier treatment than makes sense on math.SE.

Edit, this is purely anecdotal and I don't have much to back it up with but I've noticed that Europe is more interested in stochastic processes than America. I don't know why, it just seems that way. This is mostly an American site. I noticed you used << >> for quotes, so maybe your educational background is more focused on stochastic processes.

$\endgroup$
3
  • $\begingroup$ I'm from europe, but interestingly enough, in my region, not many people seem interested in stochastic processes. Here's two examples of questions I asked: math.stackexchange.com/questions/1558015/… $\endgroup$ Commented Dec 16, 2015 at 8:11
  • $\begingroup$ math.stackexchange.com/questions/1572134/… $\endgroup$ Commented Dec 16, 2015 at 8:12
  • $\begingroup$ @Anoldmaninthesea. The reason those might not be answered too much is because they require reading long technical proofs. I don't want to answer those, but I did upvote them as they are decent questions. $\endgroup$
    – user223391
    Commented Dec 16, 2015 at 8:22
3
$\begingroup$

MSE has some experts in those areas but a lot of the questions are ad hoc "practically oriented" material from areas like finance, statistics, simulation, computer programming; and posters' work projects or school exercises related to those. If a problem has some specific Markov chain written as a 4x4 matrix of numbers this is more likely to be ignored than a relatively conceptual question.

Stochastic processes also gets a lot of advanced technical questions that are difficult or would require prolonged labor to solve. Generally there is less of the happy medium, in which somebody can casually spend a few minutes without serious calculation to type out an answer and dispose of an interesting problem.

$\endgroup$
3
  • $\begingroup$ I know it's anecdotal, but from all (3) my questions related to this subject, none ad hoc, none was answered (satisfactorily or not). Maybe it's my fault, but It's not the first time that I sense that in the Math.SE community the probability(not the most basic) area is brushed aside. $\endgroup$ Commented Dec 14, 2015 at 14:29
  • $\begingroup$ @Anoldmaninthesea. Could you show me your questions? I might answer or explain why they weren't answered. $\endgroup$
    – user223391
    Commented Dec 15, 2015 at 22:40
  • $\begingroup$ @avid19 Check my questions in my math.SE profile. ;) $\endgroup$ Commented Dec 15, 2015 at 23:56

You must log in to answer this question.

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