Loading…
On the minimax decision rules in ranking problems
For the risk measured by the probability of an incorrect decision and under mild assumptions on underlying distributions, the minimax value for a general ranking problem is obtained. A necessary condition for the minimaxity is also derived. Consequently, the minimaxity of natural decision rules, for...
Saved in:
Published in: | Statistics & probability letters 1997-06, Vol.34 (2), p.179-186 |
---|---|
Main Authors: | , , |
Format: | Article |
Language: | English |
Subjects: | |
Citations: | Items that this one cites Items that cite this one |
Online Access: | Get full text |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Summary: | For the risk measured by the probability of an incorrect decision and under mild assumptions on underlying distributions, the minimax value for a general ranking problem is obtained. A necessary condition for the minimaxity is also derived. Consequently, the minimaxity of natural decision rules, for location and scale probability models, is proved. Minimaxity of natural decision rules for elliptically symmetric distributions is also studied. |
---|---|
ISSN: | 0167-7152 1879-2103 |
DOI: | 10.1016/S0167-7152(96)00180-0 |