Talk:E-statistic

Latest comment: 13 years ago by Mathstat in topic Source?

Source? edit

Does anyone have a secondary source that actually uses either term "E-statistic" or "energy-statistic"? Melcombe (talk) 14:32, 23 August 2010 (UTC)Reply

The terms "energy statistics" (plural) and "E-statistics" were introduced by G. J. Szekely to describe a class of multivariate statistics and hypothesis tests that are all based on the same basic inequality and are degenerate kernel V-statistics. There is a series of technical reports in Hungarian and later in English "E-statistics: the energy of statistical samples" by Szekely, prior to publication of several articles cited.
The term "E-statistic" is more correctly "E-statistics" because it describes a class of statistics. The original "E-statistics" was redirected to "E-statistic", which is not a term that is in use. It would be best to undo that.
Here is one description:
E-statistics (energy) tests and statistics for comparing distributions: multivariate normality, multivariate distance components and k-sample test for equal distributions, hierarchical clustering by e-distances, multivariate independence tests, distance correlation, goodness-of-fit tests. Energy-statistics concept is based on a generalization of Newton's potential energy due to Gabor J. Szekely. 24.53.137.162 (talk) 17:03, 29 October 2010 (UTC)Reply
The article has references to Szekely's papers. The question was about "a secondary source": see Wikipedia:Secondary sources. Szekely's work is obviously ungrammatical, and so is the description quoted above, so shouldn't be counted on for how the term is actually used, if it is used in English. Hence one reason for the request for secondary and hopefully tertiary sources. Wikipedia has rules for article-naming, and prefers the singular unless it is nonsense. My interpretation is that a statistic can be an E-statistic, but would not be referred to as an "E-statistics". Melcombe (talk) 10:36, 1 November 2010 (UTC)Reply
Secondary sources - see for example
article: http://eprints.whiterose.ac.uk/10328/ (energy is used in article and in keywords)
article: http://pubs.amstat.org/doi/abs/10.1198/016214506000001437 (see Sec. 4.3), doi:10.1198/016214506000001437.
dissertation abstract: http://krex.k-state.edu/dspace/bitstream/2097/590/1/GeorgevonBorries2008.pdf p. 18
article (preprint): http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.70.1169&rep=rep1&type=pdf
(see e.g. last page before Conclusions) and link to published article front page
http://ieeexplore.ieee.org/xpl/freeabs_all.jsp?arnumber=1648861
book chapter: https://mail.sssup.it/~amoneta/nptci.pdf (see p. 65) —Preceding unsigned comment added by Mathstat (talkcontribs) 18:00, 1 November 2010 (UTC)Reply