3

I would like to define an authorship of a delta method, which I need to make a proper reference in my work.

As far as I know, the earliest record is dated back to 1936 when Robert Dorfman has published his A Note on the δ-Method for Finding Variance Formulae in The Biometric Bulletin, 1, 129–137 (I have not found this article by myself, but I base this on some mention found in the internet.)

Is it true that the author of delta method is R. Dorfman or someone else? Does anybody know where to reach this particular article or other from which this method comes?

Adam Przedniczek
  • 1,134
  • 3
    For an interesting review of the history of the Delta method, see Ver Hoef, Jay M. (2012) Who Invented the Delta Method? The American Statistician, 66, 124-127. – bdeonovic Feb 20 '16 at 18:27
  • @bdeonovic Thanks, this article (by Jay M. Ver Hoef) confirms a hearsay I've found, but still I cannot reach this particular article by R. Dorfman. I thought I would be able to find it at Jstor ( https://www.jstor.org/journal/biombull ), but the very 1st volume is dated to 1945 and I'm looking for rather 1936 (how that could be possible?). Does you (or anyone else) know where I can be found, because I don't want to make a reference to an article with which I not familiar. Of course, the delta method itself I found in more contemporary book, but I would like to give a genuine authorship. – Adam Przedniczek Feb 20 '16 at 18:59
  • 1
    Dorfman might have named it but I am certain that the method was known way before the 1930s. For instance, afaik Fisher used it to derive the asymptotic distribution of the correlation coefficient. – JohnK Feb 20 '16 at 19:17
  • @JohnK So you've shed some light on it. To make this situation simpler: If you (JohnK, bdeonovic and other) were me and you're about to put a referenece to delta method, what would you do: 1. threat it like binomial theorem, which is pervasively known and don't put anything, 2. put reference to the contemporary source, which only runs off former results with in 100% inadequate author, 3. puth this R. Dorfman without even posessing this '36 article. What would you do being in my situation? – Adam Przedniczek Feb 20 '16 at 20:57
  • 2
    Either 1) or 2), probably 2) if you are in academia. Just reference a standard statistical text outlining the method, for instance Elements of Large Sample Theory by E.L. Lehmann. – JohnK Feb 20 '16 at 21:16
  • 1
    I would probably not cite anything for the delta method – bdeonovic Feb 20 '16 at 23:07

0 Answers0