Talk:Homoscedasticity

From Wikipedia, the free encyclopedia
Jump to: navigation, search
WikiProject Statistics (Rated Start-class, Low-importance)
WikiProject icon

This article is within the scope of the WikiProject Statistics, a collaborative effort to improve the coverage of statistics on Wikipedia. If you would like to participate, please visit the project page or join the discussion.

Start-Class article Start  This article has been rated as Start-Class on the quality scale.
 Low  This article has been rated as Low-importance on the importance scale.
 

Move this page to the correct spelling[edit]

J. McCulloch has an article in Econometrica (1985) showing that the term is properly spelled with a "k": http://www.ime.usp.br/~abe/lista/pdfZAptC9KazU.pdf. "skedastic" is now the overwhelming convention used in statistics and econometrics. This page should be stored there and not under this incorrect spelling.67.194.92.192 (talk) 23:40, 26 January 2012 (UTC)

Really? The idea doesn't seem to have taken root. Google scholar (.co.uk) searches, restricted to 2011–2012, give me about 4,390 results for homoscedasticity and only 953 for homoskedasticity. Professor McCulloch's brief paper appears somewhat tongue in cheek: so it would be superfluous to suggest that, like the good statistician he is, he should examine some of his assumptions. MistyMorn (talk) 15:04, 27 January 2012 (UTC)

Requested move[edit]

HomoscedasticityHomoskedasticity – The complementary concept page is spelled "Heteroskedasticity" and is significantly more extensive than this one. Since there seems to be no generally-accepted spelling, this seems as reasonable a criteria for maintaining consistency as any. (Personally, I like the way the 'c' looks, but I'm not zealous about it and am willing to follow the established trend.) I would have just moved it myself, except that I earlier went and created a redirect which is blocking the move, before thinking better of it. Arvedui 23:14, 3 October 2006 (UTC)

Survey[edit]

Add "* Support" or "* Oppose" followed by an optional one-sentence explanation, then sign your opinion with ~~~~

  • Oppose Move Heteroskedasticity instead. Just because one article cannot spell classicisizing names correctly is no reason to spread the error. Septentrionalis 05:17, 5 October 2006 (UTC)
Ok, set it up then and I'll vote in favor. Otherwise, there really doesn't seem to be a generally-accepted "correct" spelling on this score, so moving the shorter one seems the most reasonable alternate criterion. --Arvedui 22:48, 5 October 2006 (UTC)

Discussion[edit]

Etymology[edit]

What's the meaning of scedast or skedast? If it's stated that one version is prefered to the other based on ethymologies, I think there should be some etymological explanation of the root used to construct the word! Albmont 12:16, 10 November 2006 (UTC)

See the following: "A 'c' is often used instead of a 'k' in the spelling of heteroskedasticity. The research by McCulloch [1985] concludes that the word is derived from Greek roots and the proper English spelling is with a 'k'. Reference: J. Huston McCulloch, "On Heteros*edasticity", Econometrica, Vol. 53, 1985, p. 483." 75.146.224.18 (talk) 14:49, 7 February 2009 (UTC)

Shouldn't someone who actually works in statistics[edit]

Have had a voice in this, as opposed to a classics focused individual? Skedastic is the preferred spelling, as evinced on every other page relating to the subject besides the two primary ones. Ridiculous. 170.140.214.100 18:33, 2 December 2007 (UTC)

I created the initial page with this title. I have a Ph.D. in statistics. It is ridiculous to consider me a classics-focused individual. Michael Hardy (talk) 05:15, 20 August 2008 (UTC)
The spelling is absurd. It is ridiculous to believe that anyone on Wikipedia who asserts they have a PhD actually does. —Preceding unsigned comment added by 129.97.120.213 (talk) 13:45, 7 May 2011 (UTC)
Both spellings are acceptable; there's a long and honourable tradition of inconsistency in how we translate the Greek 'k' for English words. For an example from "someone who works in statistics", the Data and Statistical Services guys at Princeton are happily spelling it with a 'c'. The Random House Dictionary also spells it 'homoscedastic'. Googling will find you plenty more examples of the 'c' spelling in journal publications and dictionaries. --GenericBob (talk) 14:13, 7 May 2011 (UTC)

Bad illustration[edit]

The picture is a comic-book type of thing. It's a bounded distribution—probably uniform on a short interval. Silly. Michael Hardy (talk) 05:15, 20 August 2008 (UTC)

when compared to the similar chart in the hetero- article it's probably good enough to illustrate the concept. Matt Whyndham (talk) 16:22, 7 January 2009 (UTC)

Text related to illustration[edit]

Text states: "In a scatterplot of data, homoskedasticity looks like an oval." This seems incorrect; I would expect the y-height (proportional to s.d.) to be independent of x. Then a scatterplot of (x,y) where y=cx+noise would look like a parallelogram. An "oval" would suggest that the variance decreases as one approaches the x endpoints. Perhaps the entire paragraph should be eliminated. Comments? Rl1rl1 (talk) 17:52, 30 September 2008 (UTC)

yes quite. the text should at least say what two variables the scatter plot plots. I read "oval" too and thought of a tight locus of points, whereas we are just talking about the symmetries of the blob, no? Matt Whyndham (talk) 16:18, 7 January 2009 (UTC)

This is just a bad example. I'm used to working with scatters of two random leptokurtotic variables. This data always looks like an oval, whether it is homo- or hetero- skedastic. 75.146.224.18 (talk) 14:50, 7 February 2009 (UTC)

I also found this confusing - especially since the accompanying scatterplot illustrating homoskedasticity is not remotely oval-shaped. I'm going to remove this bit for now. --GenericBob (talk) 00:59, 15 May 2009 (UTC)

Consistency[edit]

Once the skedast vs scedast thing is worked out, the article should be editted to only use one spelling throughout (mention the alternate spelling at the beginning as is the case right now). The article right now alternates randomly bewteen and c and k spellings. Kcgrant (talk) 02:18, 25 April 2009 (UTC)

Minor Typo[edit]

You begin with a bracket at "(assuming a distribution of data ...", but then never close the bracket anywhere.Runestone1 (talk) 02:30, 12 April 2011 (UTC)