6
$\begingroup$

Why should questions contain motivation and why does the absence of motivation make questions ambiguous?

I wrote this question Probability Mass Function making the Truncated Normal Discrete and it was closed because it was ambiguous probably because it lacked motivation.

$\endgroup$
18
  • 6
    $\begingroup$ It's very difficult to advice someone on what to do if you can't figure out what they're trying to do. I would have closed that question, too. It would be better to describe your situation (perhaps a scientific study), what the data are (eg, number correct on a test), & what your goals are (eg, you need the distribution to conduct a power analysis), etc. Then we'd be much better able to help. $\endgroup$ – gung - Reinstate Monica Dec 20 '20 at 13:12
  • 5
    $\begingroup$ Many questions lacking the "why" explanation are in fact XY problems. We do not want to answer the wrong kind of question. $\endgroup$ – Tim Dec 20 '20 at 20:01
  • $\begingroup$ @Tim I already answered why. Most people when they have such data use Normal or Truncated Normal at best. I don't think the ease to manipulate a continuous distribution warrants the approximation. $\endgroup$ – George Ntoulos Dec 20 '20 at 20:51
  • $\begingroup$ @gung-ReinstateMonica I tried rewording the entire question. $\endgroup$ – George Ntoulos Dec 28 '20 at 12:40
  • $\begingroup$ @Tim I tried rewording the entire question. $\endgroup$ – George Ntoulos Dec 28 '20 at 12:40
  • $\begingroup$ I still don't know what it is about. What is the situation where this occurs? What are the data? What do you want to do with / learn from the data? What will you use the distribution for? Etc. $\endgroup$ – gung - Reinstate Monica Dec 28 '20 at 13:24
  • $\begingroup$ @gung-ReinstateMonica I want to describe the distribution (the allocation/the division of something with exactitude and methodicity i.e. as the word distribution is used in the English language and not in Statistics). What use is there in a distribution (statistical technical term) other than to describe the distribution (a common natural word in the English language)? The situation occurs any time we are describing the distribution (the English word) with Normal or Truncated Normal while knowing that the data (and the general population/distibution) is both bound from both ends and discrete. $\endgroup$ – George Ntoulos Dec 28 '20 at 21:21
  • $\begingroup$ @gung-ReinstateMonica I just asked english.stackexchange.com/questions/556068/… What I want to do with my Distribution is to Describe the Distribution of the Object of my Study. In this example that my data are Grades I want to describe the Distribution of Grades in the Class. $\endgroup$ – George Ntoulos Dec 30 '20 at 0:52
  • $\begingroup$ @gung-ReinstateMonica I once more editted the question explaining what I want to do with the Distribution (Describe the Spread). $\endgroup$ – George Ntoulos Jan 3 at 20:21
  • $\begingroup$ @gung-ReinstateMonica Isn't the question now clear and unambiguous? $\endgroup$ – George Ntoulos Jan 5 at 22:01
  • $\begingroup$ Not to me. All my questions above still hold. Maybe it's clear to someone else. $\endgroup$ – gung - Reinstate Monica Jan 6 at 6:56
  • $\begingroup$ @gung-ReinstateMonica Even the last 2 questions? What do you want to do with / learn from the data? Describe the spread in the grades of the population (including the studen't that didn't self-select). What will you use the distribution for? Describe the spread of the grades (or any other variate). $\endgroup$ – George Ntoulos Jan 6 at 11:50
  • $\begingroup$ If you want to describe the spread, try computing the SD. The question isn't clear to me; maybe it is to someone else. $\endgroup$ – gung - Reinstate Monica Jan 6 at 12:30
  • $\begingroup$ @gung-ReinstateMonica I want to use a Distribution for describing. Statistically nothing is more informative than a distribution. That is why the result in Bayesian Statistics is often the posterior in itself. It contains all the information that may be infered. It is Frequentists who ask them to give a Point Estimate or an Interval Estimate. The same SD may belong to 2 completely different distributions. The SD alone is not descriptive or informative enough. $\endgroup$ – George Ntoulos Jan 6 at 12:50
  • $\begingroup$ @gung-ReinstateMonica All your questions above still hold; why? Does the body of my main question not answer your questions (in the way I have answered them here in the comments not in a way you would like them to be answered), or is it a matter of disagreement (aesthetic maybe)? Does the body lack information I have given in the comments or do you simply disagree with using a distribution to describe the spread? Infinite distributions have the same SD and that is why I strictly want a distribution to describe the spread and not simply the SD. $\endgroup$ – George Ntoulos Jan 11 at 5:29
12
$\begingroup$

It is difficult to read what your question is about. Your introduction is:

I really like the truncated normal but I want to make it discrete. I need a discrete distribution and I hate approximations (the best fit for my data is truncated normal but my data are discrete [0,25]) I am fixated on the kernel and the nice relations and ratios of the different values.

So you hate approximations... but what is the question?

I already answered why. Most people when they have such data use Normal or Truncated Normal at best.

This doesn't clarify at all why you want to discretize the normal distribution and what the problem is.


The problem is very general/broad and cannot be easily answered without the question having more focus.

There might be a reason why you 'really like the truncated normal'. But if you leave it open for others to guess why this is the case, then this information is not very useful.

  • If you want some discretized truncated normal distribution, what are the properties of the truncated normal that should remain when the distribution is discretized?
  • How do you want the discretization to be done? The underlying mechanism in your problem, why it is that a normal distribution is a good model, will influence the way to do the discretization.

If you explain the problem and the motivation then all such considerations do not need to be guessed.


And to be honest. I did not read the second half of the question. It is a big block of text which might become more readable if you split it into multiple paragraphs.

$\endgroup$
1
  • $\begingroup$ Comments are not for extended discussion; this conversation has been moved to chat. $\endgroup$ – Sycorax Jan 12 at 4:46

You must log in to answer this question.

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