Nothing Special   »   [go: up one dir, main page]

Skewness: Difference between revisions

Content deleted Content added
Line 147:
 
Bowley's measure of skewness (from 1901),<ref>Bowley, A. L. (1901). Elements of Statistics, P.S. King & Son, Laondon. Or in a later edition: BOWLEY, AL. "Elements of Statistics, 4th Edn (New York, Charles Scribner)."(1920).</ref><ref name=Kenney1962>Kenney JF and Keeping ES (1962) ''Mathematics of Statistics, Pt. 1, 3rd ed.'', Van Nostrand, (page 102).</ref> also called '''Yule's coefficient''' (from 1912)<ref>Yule, George Udny. An introduction to the theory of statistics. C. Griffin, limited, 1912.</ref><ref>{{cite journal | last1 = Groeneveld | first1 = Richard A | year = 1991 | title = An influence function approach to describing the skewness of a distribution | journal = The American Statistician | volume = 45 | issue = 2| pages = 97–102 | doi = 10.2307/2684367 | jstor = 2684367 }}</ref> is defined as:
:<math>\frac{\frac{{{Q}(0.753/4)}+{{Q}(0.251/4)}}{2}-{{Q}(0.51/2)}}{\frac{{{Q}(0.753/4)}-{{Q}(0.251/4)}}{2}}
=\frac{{{Q}(0.753/4)}-2{{Q}(0.51/2)}+{{Q}(0.251/4)}}{{{Q}(0.753/4)}-{{Q}(0.251/4)}},</math>
where ''Q'' is the [[quantile function]] (i.e., the inverse of the [[cumulative distribution function]]. The numerator is difference between the average of the upper and lower quartiles (a measure of location) and the median (another measure of location), while the denominator is the [[semi-interquartile range]] <math>({Q}(0.753/4)}-{{Q}(0.251/4))/2</math>, which for symmetric distributions is the [[Average absolute deviation|MAD]] measure of [[statistical dispersion|dispersion]].
 
Other names for this measure are Galton's measure of skewness,<ref name=Johnson1994>{{harvp|Johnson, NL|Kotz, S|Balakrishnan, N|1994}} p. 3 and p. 40</ref> the Yule–Kendall index<ref name=Wilks1995>Wilks DS (1995) ''Statistical Methods in the Atmospheric Sciences'', p 27. Academic Press. {{isbn|0-12-751965-3}}</ref> and the quartile skewness,<ref>{{Cite web|url=http://mathworld.wolfram.com/Skewness.html|title=Skewness|last=Weisstein|first=Eric W.|website=mathworld.wolfram.com|language=en|access-date=2019-11-21}}</ref>
 
Similarly, Kelly's measure of skewness is defined as<ref>{{cite web |url=http://www.math.ruh.ac.lk/~pubudu/app5.pdf |title=Applied Statistics I: Chapter 5: Measures of skewness |author=A.W.L. Pubudu Thilan |website=University of Ruhuna |page=21}}</ref>
:<math>\frac{{{Q}(0.909/10)}-2{{Q}(0.501/2)}+{{Q}(0.1/10)}}{{{Q}(0.909/10)}-{{Q}(0.1/10)}}.</math>
 
A more general formulation of a skewness function was described by Groeneveld, R. A. and Meeden, G. (1984):<ref name=Groeneveld1984>{{Cite journal | doi = 10.2307/2987742 | last1 = Groeneveld | first1 = R.A. | last2 = Meeden | first2 = G. | year = 1984 | title = Measuring Skewness and Kurtosis | jstor = 2987742| journal = The Statistician | volume = 33 | issue = 4| pages = 391–399 }}