Abstract
Rank statistics for multivariate designs of independent random vectors of varying dimension are shown to be asymptotically normal, even in case the dimensions tend to infinity. The asymptotic variance is estimated consistently. The result extends those by various other authors and applies to simple linear rank statistics as well as to signed rank statistics which can be handled as a special case of simple linear rank statistics under dependence.
Original language | English (US) |
---|---|
Pages (from-to) | 353-378 |
Number of pages | 26 |
Journal | Journal of Statistical Planning and Inference |
Volume | 42 |
Issue number | 3 |
DOIs | |
State | Published - Jan 1 1994 |
All Science Journal Classification (ASJC) codes
- Statistics and Probability
- Statistics, Probability and Uncertainty
- Applied Mathematics