[EMAIL PROTECTED] (Dan Esperantos) wrote in message news:<[EMAIL PROTECTED]>... > > You mentioned Pearson's result. Could you post > a pointer to a book/paper which I can use and cite. > It is likely that I can not find the original from 1901, > but any recent discussion of his or related result would do. > > Also, correct me if I'm wrong, but I think, > if N -> infinity, then P -> 1 (for any K=2,3,4,...), given that p>q, > and P -> 0, if p<q.
Sorry for the delay, but I've been away. The Pearson reference is Mathematical contributions to the theory of evolution.--VII. On the correlation of characters not quantitatively measureable. Philosophical Transactions of the Royal Society of London, Series A, 195:1-47 (1901). The only reference I remember to the Pearson paper is in N.J.Castellan, On the estimation of the tetrachoric correlation coefficient. Psychometrika, 31:67-73 (1966). You might also look at the recent (14 Apr 2004) sci.stat.math thread "bivariate normal distribution (Looking for source code)", especially the last reference in beliavsky's post and the references it contains. Your conjecture is correct: as N -> infinity, P -> 1 or 0 according as p > q or p < q. . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
