|
Activity Number:
|
261
|
|
Type:
|
Invited
|
|
Date/Time:
|
Tuesday, July 31, 2007 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
IMS
|
| Abstract - #308058 |
|
Title:
|
Sufficiency and the Preservation of Fisher Information
|
|
Author(s):
|
David Pollard*+
|
|
Companies:
|
Yale University
|
|
Address:
|
Box 208290 Yale Station, New Haven, CT, 06520,
|
|
Keywords:
|
Fisher information ; DQM
|
|
Abstract:
|
The Fisher information (FI) for a set of data is never smaller than the FI for some function of the data, T(data), with equality at least when T is a sufficient statistic. Recently, Kagan and Shepp (2005 Amer. Statistician) gave an example of an insufficient statistic for which FI(data) = FI(T). This talk will explore the relationship between loss of FI and sufficiency, when expressed using Hellinger differentiability and Le Cam's theory for equivalence of experiments.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2007 program |
