5
$\begingroup$

The tag currently has the following tag wiki excerpt:

The error of an observed value is the deviation of the observed value from the (unobservable) true function value. Do NOT use this tag for SOFTWARE ERROR messages.

I am unhappy about the "unobservable". Y'all may have noticed that I have been working on tags related to point forecast accuracy. Most questions in this area also carry the tag, and the "error" in question is the one between the point forecast and the eminently observable future realization.

I propose changing the tag wiki excerpt as follows:

The error of an estimate or prediction is the deviation between this value and the "true" value. The "true" value may be unobservable (as in estimating regression parameters), or observable (as in forecasting or predicting future or yet-unknown realizations). Do NOT use this tag for SOFTWARE ERROR messages.


In addition, the tag wiki currently reads as follows:

In statistics and optimization, statistical errors and residuals are two closely related and easily confused measures of the deviation of an observed value of an element of a statistical sample from its "theoretical value". The error of an observed value is the deviation of the observed value from the (unobservable) true function value, while the residual of an observed value is the difference between the observed value and the estimated function value.

The distinction is most important in regression analysis, where it leads to the concept of studentized residuals.

Reference: Wikipedia.

To be honest, I don't see how this can be salvaged to (a) give more information than the excerpt above, but at the same time (b) address estimating / predicting both unobservables and observables.

Therefore, I propose deleting the tag wiki altogether.

(I'm open to this confusion implying that we should have two different error tags, e.g., estimation-error and forecast-error or prediction-error. However, I don't really want to retag 443 questions.)

This is related: Error tag is ambiguous

What do you think? I didn't want to change a tag with 443 questions without a discussion.

$\endgroup$
  • $\begingroup$ I think your suggested excerpt is too long, and the second sentence can be removed from it. I would also remove scare quotes around "true" in the first sentence. Just write "The error of an estimate or prediction is the deviation between this value and the true value". You can then place your full suggested excerpt into the wiki. $\endgroup$ – amoeba Apr 15 '16 at 13:54
  • $\begingroup$ Also, I would replace "Do NOT use this tag for SOFTWARE ERROR messages" with "Use [error-message] tag to ask about software errors". $\endgroup$ – amoeba Apr 15 '16 at 13:55
  • 2
    $\begingroup$ Apart from that, I would not use (the nonexisting until now) [error] tag on Meta to ask about [error] tag on the main site; why not using our generic [tags] tag instead? +1, by the way. $\endgroup$ – amoeba Apr 15 '16 at 13:57
  • $\begingroup$ I agree w/ @amoeba. I added some relevant tags. I'll let you decide if you want to delete the [error] tag. If you think it is important for meta.CV's organization & want to keep it, you should probably add a tag excerpt & wiki for it. $\endgroup$ – gung Apr 15 '16 at 14:20
  • $\begingroup$ I am not sure how relevant this is, but here is a related question pointing out that observed forecast errors (defined as differences between forecasts and realizations) may be imperfect measures of unobserved forecast errors (defined as differences between forecasts and unobserved population quantities). $\endgroup$ – Richard Hardy Apr 16 '16 at 15:44
5
$\begingroup$

To incorporate some of @amoeba's suggestions, how about:

The error of an estimate or prediction is its deviation from the true value, which may be unobservable (e.g., regression parameters), or observable (e.g., future realizations). Use the [error-message] tag to ask about software errors.

That has 234 characters instead of 308.

I would leave the full wiki as is for the moment. I think the ideas can be brought together and explained coherently. I'll have to think about it.

$\endgroup$
  • $\begingroup$ +1 but perhaps either use comma after "eg" both times or not at all :) Also, why "eg" and not "e.g."? $\endgroup$ – amoeba Apr 15 '16 at 14:31
  • $\begingroup$ I just missed it, @amoeba. In comments, I use "eg" rather than the correct "e.g." to save 2 characters. Tag excerpts are analogous to comments with respect to trying to shave characters. $\endgroup$ – gung Apr 15 '16 at 14:34
  • 1
    $\begingroup$ I see. Personally, I would prefer the version with dots in the wiki excerpts. $\endgroup$ – amoeba Apr 15 '16 at 14:38
  • $\begingroup$ +1, but I would not remove the 'Do NOT use this tag for SOFTWARE ERROR messages'. Doing the way you and @amoeba propose is better guidance about using the correct tag, but also sounds like encouraging such type of questions. $\endgroup$ – Andre Silva Apr 15 '16 at 17:48
  • 1
    $\begingroup$ This looks nice. I'll watch this for now, in case someone has an even better idea and/or you do propose a nice full wiki. (Perhaps put that in a separate answer?) I'd include the periods, though, or I'd feel tempted to also remove vowels, nd w ll knw y cn rd txt wtht vwls jst fn, rght, nd t wld sv vn mr chrctrs. $\endgroup$ – Stephan Kolassa Apr 15 '16 at 18:17
  • 2
    $\begingroup$ @Andre Unless we think that all questions referring to [error-message] are off-topic, I think it's fine to redirect people to this tag. I mean, this tag does exist. $\endgroup$ – amoeba Apr 15 '16 at 21:27
  • $\begingroup$ I have edited the tag wiki excerpt as per your suggestion and have left the wiki alone for now. Any thoughts on that? $\endgroup$ – Stephan Kolassa Apr 20 '16 at 6:29
  • $\begingroup$ @amoeba actually I'd consider whole [error-messages] tag off-topic... $\endgroup$ – Tim Apr 24 '16 at 19:08
  • $\begingroup$ Hi @StephanKolassa, just came across this old question of yours; do you still consider this issue unresolved? You did not accept gung's answer... $\endgroup$ – amoeba Apr 4 '17 at 9:17
  • 1
    $\begingroup$ @amoeba: thanks for the ping! I accepted the answer. $\endgroup$ – Stephan Kolassa Apr 4 '17 at 16:45

You must log in to answer this question.

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