Should the template include notable statisticians? I think it may be too much, but what does everybody else think? Lilac Soul ^{(talk • contribs • count)} 08:23, 28 August 2007 (UTC)

If we do, we should probably only include the top five or six or so. --Arcadian 11:47, 28 August 2007 (UTC)

It doesn't seem relevant to me. The purpose of the template is to make it easier to navigate to related topics, and I think it's unlikely anyone will want to navigate from t-test to R. A. Fisher. --Zvika 11:37, 9 October 2007 (UTC)

It sounds like a good idea to me, but perhaps it could be it's own navbox?

Could topics like Bimodal distribution and Binomial distribution be included? They seem to be already included in the ProbDistribution navbox, but that navbox is very large. Rhetth (talk) 22:41, 4 August 2008 (UTC)

Apparently, links to Wikipedia:WikiProject_Statistics in this template violates WP:SELF - WikiProjects should not be included in articles or on their templates. I have removed.—G716 <^T·_C> 01:13, 22 September 2008 (UTC)

This template has become quite large, and as WP:NAV says, "a large template has limited navigation value". I suggest splitting it into several smaller templates on each sub-topic. (We could draft them as sub-pages of this talk page to start with, e.g. Template talk:Statistics/Experimental design.) We could also consider a "Major topics in statistics" navbox to give an overview, in the same style as e.g. Template:Number theory-footer, Template:Applied-footer etc. Thoughts / volunteers? (I'm not taking this on on my own..) Qwfp (talk) 10:22, 1 October 2008 (UTC)

"Quite large" is, of course, relative. This template is bigger than a breadbox ({{Mathematics-footer}}), but not as big as a Mack truck ({{HIVpharm}}, {{Medicine}}, {{Influenza}}, {{Compression Methods}}). Personally, I think that it's ok at the moment, but I would support removing the "History" section (that I added by don't own) and the "Publications" section to keep the template from becoming list cruft. I'm not sure that some of the possible sub topics are big enough to rate a separate template. If this idea moves forward, I'll be willing to help out. Regards—G716 <^T·_C> 01:04, 2 October 2008 (UTC)

We could make it collapsible: {{Herbs & spices}}—G716 <^T·_C> 01:12, 2 October 2008 (UTC)

or collapsible, with autoshow groups: {{ProbDistributions}}—G716 <^T·_C> 05:57, 3 October 2008 (UTC)

We could use {{Least squares and regression analysis}} as one of the sub-topic templates.—G716 <^T·_C> 05:49, 3 October 2008 (UTC)

I have removed publications and history sections to reduce size of template. If anyone objects, please revert. —G716 <^T·_C> 12:20, 16 April 2009 (UTC)

Any thoughts on the relevance of the first three items under this heading: Null hypothesis • Alternative hypothesis • Type I and Type II errors • Statistical power • Effect size. I would suggest dropping them entirely as they aren't needed elsewhere in a brief list of topics. Melcombe (talk) 15:55, 9 December 2008 (UTC)

Type I & II error needs to stay, because reducing the probability of a Type II error is the usual concern for minimizing the number of replications (for ethical & economical reasons). Also, the Type I error is usually fixed; the phrasing is "Type I & II error", so keeping Type I error requires only 4 characters, "I & ". Removing the hypotheses is okay. (Duplicates comments at the ExpDes Template, addiing the character count ) Kiefer.Wolfowitz (talk) 15:45, 25 June 2009 (UTC)

On this template's talk page, I provided an example of a substantially shortened experimental design template. I commented out the topics already suggested for deletion. I commented out the RSM geometries (cube, simplex, sphere), the algebra of balanced designs (association schemes, Bose Mesner algebra, spherical designs), and the specialty entries on sequential designs (armed bandits, gittins index). I put the observational studies topics on their own line (below repeated measurements, because of similarities and the weakness of the Repeated measurements article, imho).

The option {{Statistics|state=collapsed}} appears not to work .... can it be fixed? Melcombe (talk) 10:21, 2 December 2009 (UTC)

I couldn't make sense of the collapsing features on the experimental design template. Kiefer.Wolfowitz (talk) 03:48, 18 January 2010 (UTC)

 Done I have added the "state" parameter. The template can now be collapsed in an article.

Well done! Kiefer.Wolfowitz (talk) 21:43, 20 May 2010 (UTC)

I doubt that volcano plots deserve to listed on this template, and am concerned that they were added without prior discussion. I have asked the responsible editor to consider removing this item, and to discuss the importance of such plots here before adding them. Thanks! Kiefer.Wolfowitz (talk) 22:15, 3 January 2010 (UTC)

Sorry for adding it without discussion. You're probably right that it's not widely used in statistics, but it's becoming increasingly common in genomics, etc. If you (and whoever maintains this statistics template) feel like it should be removed I understand and won't argue. Some of the other plots in that section (i.e. boxplot, bar chart) are obviously completely ubiquitous, but (to me, a non-statustician) some of the others (stem plot, q-q plot) seemed no less-obscure than volcano plots. I got the feeling that some of these are niche plotting-techniques and others are widespread. I'm happy to let you remove it at your discretion and/or let others opine. Roadnottaken (talk) 22:48, 3 January 2010 (UTC)

Thanks for your nice response. Quantile-quantile (qq-)plots are the standard for residual-diagnostics---and are apparently more effective than histograms, etc. They are the standard output in S-Plus and R. Stem-plots are widely used but not always admired (Tufte, if my memory is correct). Since you don't object, I'll remove the plot, which may be included in the future, of course, as it becomes more popular. Happy New Year! Kiefer.Wolfowitz (talk) 00:05, 4 January 2010 (UTC)

I comment on imperfections: "Non-standard" is wrong because the Gauss-Markov theorem justifies least squares.

Instead, I would suggest something like

• Linear models and generalized linear models: Regression
• Nonlinear and semiparametric models:

Thanks. Kiefer.Wolfowitz (talk) 19:00, 21 January 2010 (UTC)

I used "Non-standard" mainly as a contrast to the linear structure of the predictor in the "standard case". I think a major consideration here is not to introduce names for groupings which are too long, otherwise the template becomes too cumbersome. (I note that even the recent change of a word from "choice" to "determination" has has the effect of expanding a column width-ways and hence may have has an effect of making the template have extra lines at some resoultions.) There may even be a good reason to reduce the amount of space given over to regression and correlation, partly on the grounds that there is already a separate template for that, but also because there are a lot of other major statistical topics that do not feature at all yet. Thus we could replace the regression and correlation grouping with just a list of these major topics, such as regression, design of expereiments, multivariate analysis, time series analysis, with room for more such topics. Incidently I copied over some of the structure from here to the regression template, so the points you raised above still need to be considered there. Melcombe (talk) 10:08, 22 January 2010 (UTC)

I'd like to see a section for exact tests added to the template, including the binomial test, multinomial test, Fisher's exact test, Barnard's test, and any others I forgot. Drdan14 (talk) 01:01, 8 March 2010 (UTC)

This would raise the questions of how to group (and name the groups of ) other tests. Considering that it would be impossible to fit all 2000 statistics articles on this template it may be better just to have an entry for exact tests and entries for other similar ways of grouping tests (... as a single additional row for the time being?). It may have been suggested elsewhere, but thought could be given to having a separate "statistical tests" template, which would allow freedom for including better/more subdivisions. Melcombe (talk) 09:50, 8 March 2010 (UTC)

A separate statistical tests template sounds like a great solution. Drdan14 (talk) 23:15, 15 March 2010 (UTC)

Natural experiment and observational study---these two topics are important in discussions of empirical science and especially social sciences like economics and epidemiology. I moved both of them to the back of the experimental section, since they are usually discussed as methods that are used when randomized methods are too expensive, unethical, etc. I added "randomization", which is usually taught as a fundamental concept in statistics (Peirce, Fisher, Rao, Moore McCabe, Freedman, ASA guidelines for a first course, etc.). I moved the sampling-topics to the same section on data-collection, leaving questionaires, etc. in the social science section below. Thanks, Kiefer.Wolfowitz (talk) 14:44, 30 April 2010 (UTC)

I would suggest moving survey topics back to social-statistics. There is now duplication in social-statistics and sampling. Thanks! Kiefer.Wolfowitz (talk) 01:30, 8 May 2010 (UTC)

Since i suggested above this template had got too big 18 months ago it has very nearly doubled in size! It's now too big to fit in the browser window on my laptop screen (bigger than some) without scrolling, and that certainly limits its navigational value. If we gave it a severe pruning i can't see any easy way to suppress regrowth, so i'd like to again suggest splitting it up. Most of the existing (main, i.e. leftmost) groups would be suitable for separate navboxes. Thoughts? Volunteers? (I'll volunteer to take on Survival analysis myself this time.) Qwfp (talk) 20:57, 25 May 2010 (UTC)

I agree that the template is too long (and sometimes too wide). Thanks, Kiefer.Wolfowitz (talk) 21:02, 25 May 2010 (UTC)

Possibly the simplest way of progressing would be to follow-up one of these previous suggestions (copied from above) by G716:

• We could make it collapsible: {{Herbs & spices}}
• or collapsible, with autoshow groups: {{ProbDistributions}}

As a first step, we can try to get a better grouping at the highest level. In the prrsent structure, the names of these groups need to be quite short, but the collapsible form would allow for longer, more meaningful names it seems. I will do a little regrouping. Melcombe (talk) 09:50, 26 May 2010 (UTC)

I have done some merging of groups to try to give a reasonable number of high-level groups, each having comparable content. ... Of course, changing to a collapsible form would lose the "overview" character of seeing a whole range of topics at once, so this is an important decision .... I will place an invite to discussion on the Stats project talk page. Melcombe (talk) 10:04, 26 May 2010 (UTC)

A {{Navbox with collapsible groups}} like {{ProbDistributions}} now seems a good solution to me. In fact i can't remember why i favoured splitting over this solution last October... possibly because of the Template talk:ProbDistributions#Too non-layperson criticism of {{ProbDistributions}} after I converted that, but that criticism wouldn't apply to the groupings in this case. I don't think I'm likely to have time to convert this one at present, but I'm willing to help out with the subsequent minor change that would be necessary to each article that transcludes it to specify which group it's in. Qwfp (talk) 11:57, 27 May 2010 (UTC)

I had a go at converting it to a {{navbox with collapsible groups}} after all—much easier and quicker than i'd expected. See User:Qwfp/sandbox. —Qwfp (talk) 13:50, 3 June 2010 (UTC)

This looks very good, but is it still going to be possible to have an option where only the single bar labelled "Statistics" appears, as with the existing "Statistics|state=collapsed" form ... if so, can this be put on the documentation? Melcombe (talk) 14:35, 3 June 2010 (UTC)

It's possible now. I've added this to the documentation and tried it out at the bottom of my user page. Qwfp (talk) 16:19, 3 June 2010 (UTC)

CONGRATULATIONS on the great revision! Kiefer.Wolfowitz (talk) 16:27, 6 June 2010 (UTC)

Thanks! Was wondering if anyone would even notice... Turns out changing the template was straightforward, but altering the code on every page that transcludes it to show only the relevant group is tedious. I've done the "Descriptive Statistics" group but that was enough for now, so if anyone wants to take on one of the others.. i suspect would be easier with one of the tools such as WP:AWB. Qwfp (talk) 16:46, 6 June 2010 (UTC)

Data-Collection: Updated Kiefer.Wolfowitz (talk) 17:02, 6 June 2010 (UTC) Kiefer.Wolfowitz (talk) 17:39, 6 June 2010 (UTC)

Two suggestions, which would have the topical ordering follow the ordering of elemtentary textbooks in the USA and Sweden:

First, move "data collection" up, before "inference".

Second, have the "data analysis" follow "regression". (The "data analysis" could be renamed "multivariate dependent random variables", also.)

(This is not a big issue, imho.).

My previous re-ordering of data collection and inference was effectively reverted by editor Melcombe. No other editors commented on my previous re-ordering. New blood or fresh arguments might move the ordering towards consensus! Thanks, Kiefer.Wolfowitz (talk) 23:07, 6 June 2010 (UTC)

Done. Kiefer.Wolfowitz (talk) 23:39, 12 June 2010 (UTC)

I have un-wikilinked "study design", and hidden "regression discontinuity", on a possibly temporary basis. The object here is to see what other articles link to them, and this is effectively impossible while they are in this template. It seems to take about a week for the effect of this type of removal to feed through to all of Wikipedia's underlying cataloguing which is used for the "what link's here" tool. Melcombe (talk) 14:49, 7 July 2010 (UTC)

Well, that was informative --- was it not? Given that "regression discontinuity" (now "regression discontinuity design") seems unimportant in terms of linked articles I propose that it not be included in the template. "Study design" has been renmaed to "clinical study design" to refect its content and I am about to place this under "applications", and replace the present group name of "study design" by "design of sampling". Note that the name "sampling design" is already used for an article with a different purpose. Melcombe (talk) 09:18, 14 July 2010 (UTC)

Let me argue against hiding "regression discontinuity designs". Leading econometricians and Cook & Campbell, complain about weak observational studies and when people could do better studies with better designs, like regression discontinuity designs, etc. The first encyclopedia (Diderot & friends) wrote to enlighten people: The WP Statistics template should have a similar role, of spreading "best practice" of leading statisticians (and related statistical scientists). If it is true (and it is) that leading scientists doing observational studies recommend regression discontinuity designs, then such designs deserve a place on the template. (It may be true that we should add links to "Regression discontinuity designs" on other articles.) Best regards, Kiefer.Wolfowitz (talk) 18:46, 14 July 2010 (UTC)

Of course there is the "design of experiments" navbox template that might be better suited to this "teaching" role. And this template can't hope to cope with all 2000 statistic articles, or even with everyone's personal favourites. But let's put it back for now and take it out later if the lines becomes too confused. A problem is that the name isn't particulary informative of itself, and if truncated to just "regression discontinuity" seems to refer to something else entirely. Melcombe (talk) 09:38, 15 July 2010 (UTC)

I have put "regression discontinuity" back, but it didn't seem to belong under "uncontrolled experiments", so I moved it up since there would seem to be some control(?). But above you linked it to observational studies. Melcombe (talk) 09:48, 15 July 2010 (UTC)

I have replaced the previous group name of "Data analysis" by "Data analyses and models for particular data types", partly to reflect concerns expressed above, but also as a trial of using a longer name to see if this length is acceptable to those concerned with using up too much space. Melcombe (talk) 09:32, 14 July 2010 (UTC)

"Particular data types" would cover regression, also. The topics in this section have multivariate dependence and/or are stochastic processes. The regression topics are mostly for single-period data. Kiefer.Wolfowitz (talk) 18:46, 14 July 2010 (UTC)

Both "data analysis" and "Particular data types" would cover regression, but regression might be thought to be such an important topic that it needs its own group. I will slightly adjust the name of the group. Melcombe (talk) 09:38, 15 July 2010 (UTC)

I propose moving the line headed "Correlation" from under "Correlation and regression" to be under "descriptive statistics" as correlation seems to be more a descriptive thing rather than a modelling activity. Alternatively, since lines are mostly hidden when displayed, it could be under both group headings (possibly with slightly different content). Melcombe (talk) 09:40, 14 July 2010 (UTC)

Typically, association precedes the study of regression, so I am skeptical about this change: Maybe you could suggest a trial here, on the talk page, before implementing the change?

However, if you feel inspired, then maybe you could do something with the weak line about "categorical data" in the "descriptive statistics" section, and incorporate topics from categorical association (e.g. from Bishop, Feinberg, and Holland or from Agresti)? Thanks for asking for feedback, and for your initiative. Best regards, Kiefer.Wolfowitz (talk) 18:30, 14 July 2010 (UTC)

If we went for a two line approach I would suggest: (i) under description, and possibly in a sub-group for multivariate stuff, having the present line but omitting "confounding variable", so that both Pearson and Spearman would appear; (ii) under regression, having a line with Pearson correlation only, with partial correlation plus "coefficient of determination"(possibly appearing as R²). The contents here for regression can be kept quite simple because there is the "Least squares and regression analysis" navbox template for more detailed stuff. Melcombe (talk) 09:38, 15 July 2010 (UTC)

Following the above comment about "categorical data", I have added a "Categorical data" group under analyses. I think the group under "descriptive" was meant to be about data that had been summarised into something like contingency tables, rather than analyses of contingency tables. So I have revised the groups under "descriptive" to try to be clearer. Melcombe (talk) 11:33, 15 July 2010 (UTC)

I fleshed-out the categorical data section. Caveats: The log-linear model is linked to Poisson regression; Graphical models often handles categorical data (Sundberg, Wermuth, Lauritzen, etc.), but could well be in multivariate statistics (which is already packed full, imho) also. Thanks!  Kiefer.Wolfowitz  (Discussion) 01:01, 18 February 2011 (UTC)

Estimation is obviously an inferential procedure.

Some sampling-theory estimators are listed in the grouping of procedures for "frequentist" (sic., i.e. sampling-theory) inference, which shows some inconsistencies with the present grouping. I suggest merging the "general estimators" into the "frequentist" inference grouping. Thanks, Kiefer.Wolfowitz (talk) 17:39, 26 October 2010 (UTC)

This would be impractical because the logical follow-on would be to bring all the articles under "tests" into the same group of classical inference, which would then be too large. There might be a possibility of adding subgrouping, but then the lines will again become rarther longer, and this was a previous complaint. There seem to be no articles about standard Bayesian procedures that could be made into a sensible parallel group/subgroups under Bayesian inference.... there isn't even a worthwhile article that could be interpreted as "Bayesian inference for the normal distribution", which might be an expected thing to appear. I had already noticed the point raised in the edit-comment that the Likelihood ratio test might not count as a "specific test", but then the Wald test was already listed. The obvious third candidate for a "general test" group, if such were created, would be Lagrange multiplier test, but that doesn't exist: also the fourth, permutation test, just redirects to a not-particularly appropriate (to this template) subsection of "Resampling". And exact test doesn't have wide enough coverage to be worth including in the template. I don't think it worth adding a "general test" group unless there are more than 2 entries. So, at present, we at least have all the types of classical inference (estimation, testing) following the classical inference group, which is at least is somewhat logical and avoids complicating the template with subgroups. But are there other important topics that really should appear in a general-level statistics navigation-guide? Melcombe (talk) 09:41, 27 October 2010 (UTC)

Another thoughtful answer, this time ending with a good question! Thanks! Kiefer.Wolfowitz (talk) 09:45, 27 October 2010 (UTC)

Dear Melcombe and other editors,

Moving "McNemar's test & Cohen's kappa" from their position in "Data analyses and models for other specific data types" would leave that category solely to have stochastic processes or multivariate data: Then the category could be renamed "Multivariate & Stochastic Data". Would that be acceptable?

Best regards, Kiefer.Wolfowitz (talk) 18:58, 15 November 2010 (UTC)

There is no "association" group or subgroup. There are many formal modelling and data analysis techiques and problems that relate to categorical data that take them far outside the "descriptive" group heading that might be what you were thinking of. And I think that anyone involved in survival analysis would not be happy to have that plaved under a "multivariate and stochastic" heading. The group name "Data analyses and models for other specific data types" was intended to be open-ended so that other topics can be added as sub-groups without necessarily needing a new highest-level group to cope with it ... "Multivariate and stochastic" is likely to be too restrictive. Melcombe (talk) 10:09, 16 November 2010 (UTC)

I removed "errors and residuals" as a group heading because the article contents contain nothing at all about ANOVA etc.. Similarly, Goodness of fit contains nothing much at all about regression, but is much more general in content. And Deviance (statistics) is arguably too technical for a general-scope nav-box. After all, we can't have all 2000 stats articles appearing. But I did create a new general-regression sub-group and placed "errors and residuals" within it. Melcombe (talk) 10:43, 16 November 2010 (UTC)

Dear Melcombe and others, cannot the muse inspire one of you to provide a better heading than "Formal Data analysis" (which links to nothing)? At least "errors and residuals" provided some meaningful context, and ANOVA is an extension of the "theory of errors" from the 1800s, as I noted in my edit summary. Thanks, Kiefer.Wolfowitz (talk) 13:31, 16 November 2010 (UTC)

It seems better to have a plain text group name than to have a link to an irrelevant article. Of course we could use "Analysis of variance" as the heading, and then add in articles such as Random effects model. Melcombe (talk) 14:52, 16 November 2010 (UTC)

Analysis of variance is a generic term, which is traditionally considered in terms of the orthomodular lattice of subspaces of an inner-product space. At least since Per Martin-Lof's work of the early 1970s, there have been extensions in terms of so-called statistical models, such as analysis of deviance, redundancy, etc., which try to perform ANOVA in more general settings, with varying degrees of succes and failure. Before Melcombe reverts again, would he be so kind as to specify one most reliable textbook or invited survey article that organizes this material as "formal analysis" rather than ANOVA.  Kiefer.Wolfowitz  (Discussion) 15:06, 8 April 2011 (UTC)

For the record, Melcombe replied in an edit summary, correctly noting that ANOVA is a special case of ANCOVA or Manova, so it would be illogical to have the special case include the genera! Replying, I argued that ANOVA was used also as a generic term, with less success than I'd hoped. Then I suggested Harmonic analysis, and then improved it to Euclidean analysis.  Kiefer.Wolfowitz  (Discussion) 10:32, 11 April 2011 (UTC)

Under the category of correlation and regression analysis, the navbox now has "Euclidean analysis", with links to inner-product space and to the Pythagorean identity. The interpretation of Anova as Euclidean analysis is well established in intermediate level statistics (e.g., Bailey, Bapat, Christensen, Saville & coauthor) and may be conventional: it at least provides more context and content than "formal analysis", the previous term. This resolves a minor problem that has been discussed for months, at least for me. Thanks!  Kiefer.Wolfowitz  (Discussion)

Per WP:Deviations, we should not be overriding the MediaWiki:Common.css classes using "style attributes". I attempted to remove the |belowstyle=, but was reverted. Specifying this coloring seems entirely pointless since "CCCCFF" is the default used by MediaWiki:Common.css, and hence, there is no reason to specify it. Please voice your objections here. Thank you. Frietjes (talk) 23:26, 25 April 2011 (UTC)

An editor changed this navbox (along with many others) without getting consensus or asking for comments. This is the core navbox for the statistics project, and major changes should be discussed first, here, with notice given at the project's talk page.

It is bad form, especially for a relatively inexperienced editor, to initiate changes on many navboxes, without discussing things first. There is a talk page for navboxes, and major changes to ~many navboxes generally should be made there first.  Kiefer.Wolfowitz 23:31, 25 April 2011 (UTC)

I commented out the image. I feel as though it was taking up too much space, and is not necessary here. Thank you. Frietjes (talk) 15:29, 23 May 2011 (UTC)

That's fine. I thought that having some kind of image would draw attention to statistics, but it shouldn't displace content.  Kiefer.Wolfowitz 15:54, 23 May 2011 (UTC)

I've just added a link to scan statistic under Statistical inference > statistical theory. Do feel free to move if you think it would be better elsewhere. Gareth Jones (talk) 18:08, 6 January 2014 (UTC)

I removed it, because it does not seem to be a basic topic in statistics. 184.164.133.46 (talk) 02:27, 4 August 2016 (UTC)

I think some of causal inference articles in Category:Causal inference such as Rubin causal model can be included in this template. Taha (talk) 07:04, 14 February 2014 (UTC)

I've replaced the link to Statistical population in the Applications/Social statistics section with one to Population statistics, which is the topic that would be relevant to that heading. The article Statistical population, which refers to the population of objects a statistical analysis studies, should probably be linked somewhere in the template; I'm think maybe in Data collection/Study design, or maybe in Statistical inference/Statistical theory. Anyone have a preference? -Bryanrutherford0 (talk) 18:26, 11 March 2016 (UTC)

A complaint about the overuse and size (14 Kb) of this template appeared in WikiProject Mathematics recently. Statistics is often/usually a separate department and major in American universities, and we don't use a template for history, or for large subfields like American history (see the Missouri Compromise article for how specific their infobox is), and for math the infoboxes are usually subfields like Template:topology or Template:analysis-footer.

I suggest splitting each subsection in the Statistics template to a separate footer or infobox, using those smaller templates in any lower-importance article, while a simplified general Statistics template can still be used for the high-importance general articles. SamuelRiv (talk) 16:34, 25 August 2016 (UTC)

It seems like a fairly straightforward application of Statistics which is incredibly common in the 21st century and is used in every industry. Particularly image recognition, spatial analysis, deep learning, content personalization, and so many other things... 32.97.110.60 (talk) 20:36, 20 June 2017 (UTC)

This edit request has been answered. Set the |answered= or |ans= parameter to no to reactivate your request.

Add Van der Waerden test to the "Rank statistics sub-subsection of the "Specific tests" subsection of the "Statistical inference" section of the category. B2TF (talk) 20:40, 7 March 2022 (UTC) B2TF (talk) 20:40, 7 March 2022 (UTC)

 Done Shellwood (talk) 18:52, 9 March 2022 (UTC)

add vector generalized linear model 68.134.243.51 (talk) 00:19, 22 September 2022 (UTC)

Vector generalized linear model - Wikipedia 68.134.243.51 (talk) 00:20, 22 September 2022 (UTC)

Can we add the Equivalence Test page to the section of "Testing Hypotheses". I think it fits in line with the rest of the subjects. I think it's important to include more hypothesis testing that requires a threshold for practical differences. Treodeo0 (talk) 23:41, 20 November 2022 (UTC)