Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
10525772 | Statistics & Probability Letters | 2013 | 9 Pages |
Abstract
Suppose we observe Yjn=μjn+Ïjnej for 1â¤jâ¤n in R, where {ej} are independent and identical random errors with common distribution function F(x). Let Mn=max1â¤jâ¤nYjn. When the upper tail of F is of power-type, local power type, gamma type and normal type, we give conditions on the growth of the location and scale trends {μjn,Ïjn} such that for certain constants an and bn>0, bnMnâan converges to one of the three standard extreme value distributions. In each case bn is proportional to the Lp-norm of {Ïjn} and does not depend on {μjn}. Most importantly, trend in scale is shown to dominate trend in location.
Related Topics
Physical Sciences and Engineering
Mathematics
Statistics and Probability
Authors
Christopher S. Withers, Saralees Nadarajah,