11

Given the ambiguity of empirical, I propose we get an [empirical-bayes] tag, and then we can retag all those questions pertaining the subject (which I volunteer to do as well). The other questions tagged empirical would have it removed. EDIT (4th August): We also have a tag ecdf. Some questions deserve this one, so let's try to keep it in mind.


11

Like @Scortchi and @whuber, I prefer shorter excerpts. (Besides aesthetic preference, I suspect they are more likely to be read.) I think excerpts should mostly proffer tag usage guidance, and that more substantive and detailed information is better placed in the full wiki. That said, what is "shorter" exactly? To some degree "shortness" is in the eye ...


9

I mentioned this once before, a long time ago, but can't find it now. There is [supposed to be] a distinction between them. [normal-distribution] is about the distribution itself, whereas [normality] is a property. For example, How is the chi-squared distribution related to the normal distribution? would be a question appropriate for the [normal-...


8

My opinion is that machine-learning tag is borderline useless because it is too unspecific and too vacuous. It is not the only tag like that. For example we have fitting×600 and model×900, which are also vacuous and could well be burninated & blacklisted as far as I am concerned. But machine-learning with over 12k threads is the most prominent example of ...


7

I wrote the following wiki excerpt: Stochastic gradient descent (SGD) is a variant of gradient descent where only a small subset ("mini-batch") of training examples is used to compute the gradient on each iteration. Feel free to improve. It might make sense to create a synonym [stochastic-gradient-descent]$\to$sgd. Update: done.


7

I'd go for the more concise the better, to increase the chance people actually read the tag wiki excerpt before using the tag & perhaps therefore use it correctly. The usage guidance, or tag wiki excerpt, is a short blurb that describes when and why a tag should be used on this site specifically. Mousing over a tag & reading the excerpt ...


7

It seems to me that links are best not included in excerpts (of course, they do make sense in a full wiki). I think they should not be included for the following reasons: Links don't work in excerpts. For example, you cannot click on them to navigate to the indicated page. They are hard to read. It would be cleaner to put, 'See the corresponding ...


6

It's tricky but doable. Go to http://stats.stackexchange.com/tags and type "arma". You will see the original arma tag with its own question count and excerpt (this is because it was never merged into arima). There is an "edit" link there too, leading to http://stats.stackexchange.com/edit-tag-wiki/830 -- and this gets you where you want.


6

The purpose of a Wiki excerpt is not to supply as full and complete a definition as you can. It is to help people apply tags appropriately. To be effective, the excerpt therefore should: Be as short as possible, for otherwise it will be ignored. Be clear about what the tag means--but only to the point of distinguishing it from other possible tags that ...


6

That makes sense to me. Even if someone could up with some way to distinguish these, I think there would be a ton of confusion about which to use where and a lot of error in their use. So I favor making one the synonym of the other.


5

Thank you for the good discussion so far, and thanks to @amoeba for pointing me to it! I agree with many points raised so far and just want to share a few personal experiences with this topic. General scope: I always interpreted the tag [zero-inflation] as being about the phenomenon along with various models for it. Hence, I didn't expect only questions ...


5

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 ...


5

This is a bad wiki excerpt; it is imprecise and too lengthy (see What is the optimal length & level of detail of tag wiki excerpts?) Wikipedia defines algorithm as "an unambiguous specification of how to solve a class of problems". This sounds good to me. I would replace the wiki excerpt by An unambiguous specification of how to solve a class of ...


4

I will post my own answer so that we could have some sort of voting here. I am obviously for detailed excerpts. After my post, @whuber edited all excerpts that I mentioned to shorten them drastically. Here are the edits so that people can compare which version they prefer: elastic-net: https://stats.stackexchange.com/posts/184175/revisions total-least-...


4

Q&A sites like Cross-Validated are not aggregators of information available over the web. They exist because general sources like Wikipedia do not suffice, and they are supposed to offer expertise focused on their field. Obviously what we write in this forum is not parthenogenesis, but neither is just aggregation : it is a purposeful synthesis of ...


4

I for one think something (accurate) is better than nothing, so it should make a difference whether something's there yet. However, a counterargument: people might feel less welcome to change something short that's already there, even if it's incomplete. The accuracy of the information should matter more than the source (all quotation and attribution being ...


4

I have manually retagged 2 Qs with statistics tag and 1 additional Q with probability into on-topic. The remaining 8 threads with statistics are all about site statistics, so it can now be renamed into site-statistics via moderator tools. I suggest you can do it now! The wiki excerpt says Questions about sizes and other quantities of Cross Validated ...


4

The bivariate tag is a proper subset of joint-distribution. I have updated wiki excerpts as follows: Joint probability distribution of several random variables gives the probability that all of them simultaneously lie in a particular region. and Joint probability distribution of two variables. Briefly looking at the questions, it seems that most ...


4

About references: I included a link to the most upvoted answer in Should all "book recommendation" questions be classified as community wiki? in the tag wiki as additional guidance. @amoeba edited the tag excerpt to also advise using specific tags: Questions seeking external references (books, papers, etc.) about a particular subject. Always ...


4

I usually agree with @gung on tag issues, but this time I am not convinced by his answer. I want to make two points. I don't see how we can meaningfully separate tags for zero-inflated count models, hurdle count models, and zero-inflated-aka-hurdle continuous models. Because: (A) Zero-inflated and hurdle count models are clearly distinct but very similar ...


4

It is a good practice for the stackexchange communities to explain the tags by how they are used and not by what the specific term in the tag means. See for instance the points 2 to 4 in the list in this blog post from Jeff Atwood Tags are the de-facto map of allowed (and implicitly disallowed by omission) topics on your site.  ... Avoid generically ...


3

This seems really obvious. I have made [parameter-optimization] a synonym of [optimization], and I merged them.


3

I propose: We create a new tag [sensitivity-specificity], as @amoeba suggested. We make the existing tags, sensitivity and specificity synonyms of this and merge them. (We will first need to ensure that the threats tagged with sensitivity are correct, of course.) Note that we have a tag for precision-recall, so combining has some precedent. We create ...


3

I tried to modify the current excerpt in the general direction outlined by @AdamO, but keeping it easy to grasp. My understanding of what "sensitivity analysis" is, follows @gung's answer. Auxiliary methods intended to check if the outcome of some statistical analysis strongly depends on the model assumptions, preprocessing steps, presence of outliers, ...


3

In the meantime, this was extensively discussed in the answers and comments in Tags around categorical data coding / representation and the outcome was as follows: Master tag categorical-encoding has been created, and has the following three synonyms: [dummy-variables], [indicator-variables], and [one-hot]. It currently has 67 threads. Wiki excerpt: ...


3

I think it's a great idea, and we could make [statistics] a synonym. If there's some confusion about which site is referenced, you could use [main-sites-statistics] instead.


3

Some time has passed and it was possible to observe how the above cited tags evolved. After a more carefully reading I came up with this proposal: The CV tag questions (x8, +2) seems to label threads similarly to the main Stack Exchange site. Here it is their tag excerpt: For questions regarding the nature of a "question" on StackExchange sites; e.g. ...


3

This Q was mentioned in a recent discussion and I see that it is unanswered, so I thought I would answer. What can be done when the tag name represents both a function of a statistical software, but its name also suggests a broader concept of usage? In general, this is a suboptimal situation. I would try to de-ambiguate the tag. If this is not possible ...


2

I didn't really find this thread on SO useful. Maybe the feature has been discontinued and/or because the thread is outdated (2014). However, I have found a temporary solution, which might help address the problem for the time being. As @amoeba has mentioned, this is a very common problem, and displaying tag excerpts would really help people while editing ...


2

I would say sensitivity analysis is... a set of non-primary statistical analyses intended to verify the accuracy and generalizability of estimates and inference derived from a primary analysis and to make corrections where appropriate.


Only top voted, non community-wiki answers of a minimum length are eligible