Tail hypotheses in the signal plus noise model
Based on an infinite sequence of observations Yn=Xn+an, n=1,2,... with independent identically distributed random variables X1,X2,... with known distribution representing noise and constants a1,a2,... representing signal, it is impossible to distinguish with zero error probabilities the class of signals with infinite power [short parallel]a[short parallel]2=a12+a22+... from the noise (a=0).
Year of publication: |
1999
|
---|---|
Authors: | Kagan, Abram ; Shepp, Lawrence A. |
Published in: |
Statistics & Probability Letters. - Elsevier, ISSN 0167-7152. - Vol. 43.1999, 3, p. 317-319
|
Publisher: |
Elsevier |
Subject: | Tail hypothesis Zero error probability |
Saved in:
Saved in favorites
Similar items by person
-
A Sufficiency Paradox: An Insufficient Statistic Preserving the Fisher Information
Kagan, Abram, (2005)
-
Some linear models are necessarily parametric
Kagan, Abram, (1998)
-
Kagan, Abram, (1998)
- More ...