4
$\begingroup$

I find the Wiki and its excerpt of the tag in need of improvement. I could have tried to edit it myself, but I thought I better share it here and perhaps attract contributions from some more knowledgeable user(s) .


The excerpt reads:

Prediction is concerned with assessing the probability of unknown values from known values and inferred relationships.

In my opinion, this is too narrow (probabilities are not always used) and too vague (distinction from estimation not clear enough; relation to forecasting not mentioned). See more in the comments in this thread.


The rest of the Wiki text extensively uses examples of forecasting without noting that there is a separate tag that is actually more popular than the prediction tag. E.g.

Prediction involving statistical models is used in many real-world situations as for instance in forecasting electricity demand stock prices, election or sports outcomes, clinical trials, computer science, meteorology and more. Prediction isn't only about the future. Predictions can be made for unknown concurrent or past values on the basis of relevant information that is known.

Examples of prediction models are again forecasting models:

Examples of well known prediction models are vector autoregressive models or autoregressive moving averages, but many statistical models (such as simple linear regression) can be used for predictions as well.

If forecasting is a subset of prediction, then the examples are fine, but they should be introduced appropriately (i.e. first tell that a prominent subset of prediction is forecasting, and only then list the examples).

Further we have some formatting issues and a citation that focuses mostly on estimation (rather than prediction) where we estimate known (???) parameters:

Uncertainty in data, processes or parameters means that there will be uncertainty in conclusions. Statisticians call this drawing of conclusions in the presence of uncertainty, statistical inference (or just inference); in this book, inferences will be either estimation of fixed but known parameters, or prediction of unknown random quantities. (Notice that "forecasting," namely concluding something about the future, is a special case of "prediction.")

Instead of the latter paragraph I think we should make a clearer distinction between prediction, forecasting (and potentially estimation).


By the way, the same problematic citation appears in the Wiki of the . That one should also be fixed...

$\endgroup$
  • 2
    $\begingroup$ Your 1st wiki quote does discuss prediction as distinct from forecasting & explains that it is a superset of which F is a subset. The quote at the bottom (from the book) is a copy error this Google books search shows that it should be "unknown parameters". I'll fix that. $\endgroup$ – gung May 18 '17 at 12:32
  • 1
    $\begingroup$ The first quote is way too indirect about prediction vs. forecasting. What you "see" in it is not what I "read" in it. Taken literally, the connection is just not there. For a newbie, the appearance of forecasting as in bold in the second line of that quote should be confusing. $\endgroup$ – Richard Hardy May 18 '17 at 12:45
  • $\begingroup$ Hi Richard! Have you seen my answer? What do you think? $\endgroup$ – amoeba May 27 '17 at 9:23
  • 2
    $\begingroup$ @amoeba, I have made one more small correction, and in general I like the new tag Wikis and excerpts much better. (Hopefully my edit will get accepted; I do not have the necessary reputation yet to do it freely.) Then the problem is solved. $\endgroup$ – Richard Hardy Jun 14 '17 at 8:04
  • $\begingroup$ Cool. I approved your correction but there is one more vote needed. $\endgroup$ – amoeba Jun 14 '17 at 9:47
4
$\begingroup$

I like the definitions quoted in @Sycorax'es answer here: https://stats.meta.stackexchange.com/a/3078. Can we adapt them for wiki excerpts? E.g.:

prediction
Prediction of unknown random quantities, using a statistical model.

forecasting
Prediction of the future events. It is a special case of [prediction], in the context of [time-series].

Regarding the wikis, I think after @gung fixed the typo, the [forecasting] wiki is fine, and in the [prediction] wiki we could simply delete the first three paragraphs (the quote IMHO provides enough context). Would this resolve the issue?

$\endgroup$
  • 1
    $\begingroup$ UPDATE: Two days after posting this answer, I went ahead and edited wikis+experts as suggested here. Richard, let me know if this resolved the issue. $\endgroup$ – amoeba May 24 '17 at 14:48
  • 1
    $\begingroup$ Hmmm, I guess I should have moved faster. Oh well. $\endgroup$ – gung May 24 '17 at 16:31

You must log in to answer this question.

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