A problem of minimax estimation with directional information
This problem is in the area of minimax selection of experiments. Nature chooses a number [theta] in the closed interval [-1, 1]. The statistician chooses a number y in the same interval (an experiment) and is informed whether [theta] < y, [theta] = y, or 0> y. Based on this information, the statistician then estimates [theta] with squared error loss. The minimax solution of this problem is found. The minimax value is 1/(2e). The least favorable distribution involves the truncated t-distribution with two degrees of freedom. The minimax choice of experiment involves the truncated t-distribution with zero degrees of freedom.
Year of publication: |
1996
|
---|---|
Authors: | Ferguson, Thomas S. |
Published in: |
Statistics & Probability Letters. - Elsevier, ISSN 0167-7152. - Vol. 26.1996, 3, p. 205-211
|
Publisher: |
Elsevier |
Keywords: | Selection of experiments Optimal design Search games Minimum variance partitions |
Saved in:
Online Resource
Saved in favorites
Similar items by person
-
Minimax estimation of a variance
Ferguson, Thomas S., (1994)
-
Ferguson, Thomas S., (2000)
-
Ferguson, Thomas S., (2000)
- More ...