10
$\begingroup$

As a machine learner coming from computer science, I do not had a solid education in statistics. Frankly, my past data analysis experience is dominated by hands-on approaches. I have a solid knowledge about a bunch of algorithms, both ml and statistics (when to apply, properties which may affect the result etc.) , but my theoretical knowledge really has big gaps (I keep learning). Hence I am able to answer questions more in the sense of "from my experience" but not in the sense of "I know that xy is definitely true".

I will remain silent for most of the time, but sometimes certain question remain open, which I have the feeling I can provide some hints (e.g. this one: Averaged continuous Kernel Density Estimates in lieu of a discrete Kernel Density Estimate in Monte Carlo Proceedure).

Now my question: Should I please keep my mouth shut to not confuse the questioner or am I allowed to contribute to the best of my knowledge.

I apologize if this question itself is not appropriate for this site.

$\endgroup$
12
$\begingroup$

I suspect most of us struggle with this quandary most of the time. When you can provide solid theoretical support for your response, of course that's good. But even when you cannot, yet you are sharing experience, that can be just as valuable (or even more so), so please don't hesitate to contribute on that account. (It helps to indicate what kinds and amounts of experience you have.) I see this as involving some personal risk because it's possible that the experience I share could be based on a mistaken understanding or inferior approach and someone might point that out. (This happens to me all the time on some other forums, I'm afraid.) Well, so be it: it's humbling but it's a great learning opportunity!

$\endgroup$
  • 3
    $\begingroup$ Agreed! The whole point of the StackExchange sites is to collect a wide range of answers and then sort them using the community's collective expertise. There's no better way to learn than to put your knowledge up for public review. $\endgroup$ – Matt Parker Nov 19 '10 at 22:51
  • 1
    $\begingroup$ Thank you for your encouraging words ! $\endgroup$ – steffen Nov 21 '10 at 10:59
  • 1
    $\begingroup$ Fully agree, I would say that taking the risk of a weak answer is valuable as long as strong answer come out from time to time. Frenquency of strong answer do not need to be large for the community because of the voting scheme. However, for the representation people make out of you (appart from your "rep" I would say) the frequency of strong answers has to be significant enough to dominate weak answers, at least in people's mind (and strong has often more weight than weak). $\endgroup$ – robin girard Dec 15 '10 at 7:57
6
$\begingroup$

I do not see anything inappropriate in answering a question as long as it is relevant to the question at hand. If it is incorrect in some way it will be pointed out in the comments and that would enrich the experience for everyone (OP, the answerer and everyone else).

Of course, be prepared to get your ego to be bruised a bit (via downvotes, corrections etc) but that is the price you pay for learning something new on the main site. I myself have given incorrect answers and edited them where I could to provide the correct answer.

$\endgroup$
  • $\begingroup$ Also thank you ! This is a price I am willing to pay. $\endgroup$ – steffen Nov 21 '10 at 11:01

You must log in to answer this question.

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