12.07.2015 Views

1 Studies in the History of Statistics and Probability ... - Sheynin, Oscar

1 Studies in the History of Statistics and Probability ... - Sheynin, Oscar

1 Studies in the History of Statistics and Probability ... - Sheynin, Oscar

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

eliability <strong>and</strong> queu<strong>in</strong>g are known to apply ra<strong>the</strong>r complicatedanalytical models. A certa<strong>in</strong> disproportion <strong>in</strong> <strong>the</strong> current development<strong>of</strong> probability <strong>the</strong>ory consists <strong>in</strong> that a ra<strong>the</strong>r large number <strong>of</strong><strong>the</strong>oretical models (even analytically studied to a sufficient extent) iscollected, but at <strong>the</strong> same time <strong>in</strong> many cases <strong>the</strong>y were neverpractically compared with reality. Of course, a creation <strong>of</strong> a <strong>the</strong>oreticalmodel marks a necessary <strong>in</strong>itial period without which no suchcomparison, <strong>and</strong> no underst<strong>and</strong><strong>in</strong>g <strong>of</strong> <strong>the</strong> actual data is at all possible.However, too <strong>of</strong>ten a study stops at that period. At <strong>the</strong> same time, eachcomparison with reality usually calls <strong>in</strong>to be<strong>in</strong>g new models, that is,acts refresh<strong>in</strong>g <strong>in</strong> that sense as well.Figures <strong>and</strong> TablesI did not reproduce <strong>the</strong>m. Fig. 1 − 3 <strong>and</strong> Tables 1 <strong>and</strong> 2 concerned<strong>the</strong> papers <strong>of</strong> Ermolaeva (1939) <strong>and</strong> En<strong>in</strong> (1939). Tables 3 <strong>and</strong> 4 from§ 3.2 expla<strong>in</strong>ed <strong>the</strong> Fram<strong>in</strong>gham experiment. Table 3 provided <strong>the</strong>expected <strong>and</strong> actual number <strong>of</strong> taken ill, <strong>in</strong> each expected <strong>in</strong>terval <strong>of</strong>risk, separately for men <strong>and</strong> women. Table 4 showed <strong>the</strong> estimates <strong>of</strong><strong>the</strong> coefficients <strong>of</strong> <strong>the</strong> factors <strong>of</strong> risk.Notes1. See [i, Note 4]. O. S.2. This was a feature <strong>of</strong> Soviet publications (perhaps <strong>of</strong> Russian papers even now).The late Pr<strong>of</strong>essor Truesdell told me that he was unable to read <strong>the</strong>m <strong>in</strong> translation(also because translations are usually quite formal. O. S.3. Strangely enough, no editor is mentioned <strong>in</strong> <strong>the</strong> booklet. O. S.4. A list is 24 pages typescript or 16 pages <strong>of</strong> published text. O. S.5. No wonder Laplace was elected member <strong>of</strong> <strong>the</strong> French Academy (not to beconfused with <strong>the</strong> Paris Academy whose member he also was) devoted to <strong>the</strong> study<strong>of</strong> <strong>the</strong> French language. O. S.6. Some scientists (Chebyshev, Markov) did not have any superstructure. O. S.7. The author referred (not <strong>in</strong> all necessary cases) to <strong>the</strong> text <strong>of</strong> <strong>the</strong> TAP aspublished <strong>in</strong> 1886. Instead, I provided references to its English translation(2005/2009). O. S.8. This is my quotation from Laplace (2005/2009, p. 97) <strong>in</strong>serted <strong>in</strong>stead <strong>of</strong> <strong>the</strong>author’s description. O. S.9. There are mistakes. One <strong>of</strong> <strong>the</strong>m, noticed by Pearson, concerned his model <strong>of</strong>births <strong>and</strong> deaths, see Sheyn<strong>in</strong> (1976, p. 160). Then, he had been keep<strong>in</strong>g to his ownpractically useless <strong>the</strong>ory <strong>of</strong> errors <strong>and</strong> thus caused French authors to shun Gauss(Sheyn<strong>in</strong> 1977, pp. 52 – 54). Laplace’s astonish<strong>in</strong>g mistake (1796/1884, p. 504) wasto state that <strong>the</strong> planets moved along elliptical paths not <strong>in</strong> accordance with Newton’sdiscovery, but because <strong>of</strong> small differences <strong>in</strong> densities <strong>and</strong> <strong>in</strong> temperatures <strong>of</strong> <strong>the</strong>irvarious parts. O. S.10. Concern<strong>in</strong>g <strong>the</strong> precision <strong>of</strong> his estimate, Laplace (2005/2009, pp. 46 – 47)stated: after a century <strong>of</strong> new observations [...] exam<strong>in</strong>ed <strong>in</strong> <strong>the</strong> same way [...]. Seealso Cournot (1843, § 137). O. S.11. Once more, see [i, Note 4]. O. S.12. See Note 10. O. S.13. The classical set-<strong>the</strong>oretic axiomatization is thus called <strong>the</strong> first one.Gnedenko (1969, p. 118), <strong>in</strong> a brief survey <strong>of</strong> <strong>the</strong> history <strong>of</strong> this problem, named onlyone pert<strong>in</strong>ent publication, Lomnicki (1923). O. S.14. Concern<strong>in</strong>g Ville see, for example, Shafer & Vovk (2001, pp. 48 – 50). Theo<strong>the</strong>r reference is Postnikov (1960) O. S.15. I have found this translation <strong>in</strong> Google with a reference to a commentator <strong>of</strong>Planck. O. S.16. Surely Tolstoi knew about such most actively work<strong>in</strong>g scholars as for exampleMendeleev or Chebyshev. O. S.119

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!