18
$\begingroup$

The is always a disputable thing, especially within "pure statisticians". I personally think they are inevitable connected.
In a practical view, it would generate a lot of questions, and making a separate SE for it will drive to a total chaos.

Edit: Nevertheless a war has started, so I think it is time to revive the discussion.

$\endgroup$
  • $\begingroup$ Where is the war? Aren't there two, quite similar, and highly upvoted answers? $\endgroup$ – Peter Smit Jul 25 '10 at 15:28
  • $\begingroup$ About cross-validation, on the main site. $\endgroup$ – user88 Jul 25 '10 at 17:24
23
$\begingroup$

The original site description in my proposal was "For statistics, data analysis, data mining and data visualization". I specifically wanted to include data mining/machine learning as it makes no sense to separate out these methods from the more statistical methods that are usually based on stochastic models. In fact, I like to call the whole area "Data Science".

$\endgroup$
18
$\begingroup$

I think that it should be included. For instance, "Modern Applied Statistics with S" (MASS) includes all the machine learning techniques (including neural networks and SVM).

Neural networks are one of the most popular non-linear models in statistics now-a-days.

$\endgroup$

You must log in to answer this question.