|
Activity Number:
|
663
- New Developments in Modern Statistical Estimation Theory
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Thursday, August 3, 2017 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
IMS
|
|
Abstract #324789
|
View Presentation
|
|
Title:
|
Estimation and Inference of Bounded Normal Mean
|
|
Author(s):
|
Yong Seok Park* and Jeremy M. G. Taylor
|
|
Companies:
|
and University of Michigan
|
|
Keywords:
|
Normal distribution ;
Bounded parameter ;
restricted parameter space ;
confidence interval ;
frequentist statistics
|
|
Abstract:
|
Estimating parameters within restricted parameter space has been frequently occurred in science but the estimation and inference with good properties is challenging. There are two major challenges: first is exact coverage for all parameters in the restricted space and second is to avoid unrealistic optimistic confidence bound when the observation is out of parameter space. In this paper, we propose an new method to obtain exact confidence interval for bounded normal. In addition, we also propose parameter estimation and statistical test for the bounded parameter.
|
Authors who are presenting talks have a * after their name.
