This is the program for the 2010 Joint Statistical Meetings in Vancouver, British Columbia.
Abstract Details
Activity Number:
|
189
|
Type:
|
Invited
|
Date/Time:
|
Monday, August 2, 2010 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Section on Statistical Learning and Data Mining
|
Abstract - #307333 |
Title:
|
Adaptive Confidence Intervals for the Test Error in Classification
|
Author(s):
|
Eric B. Laber*+
|
Companies:
|
University of Michigan
|
Address:
|
1085 S. University, Ann Arbor, MI, 48109,
|
Keywords:
|
Classification ;
Test Error ;
Non-Regular Asymptotics
|
Abstract:
|
The estimated test error of a learned classifier is the most commonly reported measure of classifier performance. However, constructing a high quality point estimator of the test error has proved to be very difficult. Furthermore, common interval estimators (e.g. confidence intervals) are based on the point estimator of the test error and thus inherit all the difficulties associated with the point estimation problem. As a result, these confidence intervals do not reliably deliver nominal coverage. In contrast we directly construct the confidence interval by use of smooth data-dependent upper and lower bounds on the test error. We prove that for linear classifiers, the proposed confidence interval automatically adapts to the non-smoothness of the test error, is consistent under fixed and local alternatives, and does
|
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 2010 program
|
2010 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Continuing Education program, please contact the Education Department.