Percentage Points of the Largest Among Student's T Random Variable |
| |
Authors: | Mukhopadhyay Nitis Aoshima Makoto |
| |
Institution: | (1) Department of Statistics, University of Connecticut, Storrs, CT, 06269-4120, U.S.A;(2) Institute of Mathematics, University of Tsukuba, Ibaraki, 305-8571, Japan |
| |
Abstract: | Let us consider k( 2) independent random variables U1, . . . ,Uk where Ui is distributed as the Student's t random variable with a degree of freedom mi, i=1, . . . ,k. Here, m1, . . . ,mk are arbitrary positive integers. We denote m=(m1, . . . ,mk) and Uk:k=max {U1, . . . ,Uk}, the largest Student's t random variable. Having fixed 0< <1, let a a(k, ) and hm hm (k, ) be two positive numbers for which we can claim that (i) k(a)– k(–a)=1– , and (ii) P{–hm Uk:k hm}=1– . Then, we proceed to derive a Cornish–Fisher expansion (Theorem 3.1) of the percentage point hm. This expansion involves a as well as expressions such as i=1
k mi
–1, i=1
kmi
–2, and i=1
k mi
–3. The corresponding approximation of hm is shown to be remarkably accurate even when k or m1, . . . ,mk are not very large. |
| |
Keywords: | largest t value percentage point Cornish– Fisher expansion approximation applications |
本文献已被 SpringerLink 等数据库收录! |
|