|
Activity Number:
|
150
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 12, 2002 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Physical & Engineering Sciences*
|
| Abstract - #301036 |
|
Title:
|
Bayesian Density and Reliability Estimation Based on Rank Set Samples
|
|
Author(s):
|
Kaushik Ghosh*+ and Ram Tiwari
|
|
Affiliation(s):
|
George Washington University and University of North Carolina, Charlotte
|
|
Address:
|
2201 G Street NW, Suite 315, Washington, District of Columbia, 20052, USA
|
|
Keywords:
|
Ranked set samples ; dirichlet process prior ; Reliability ; Density estimation
|
|
Abstract:
|
Ranked set sampling is used to acquire data when sampling units are easily ranked based on their attributes, but exact measurements are expensive to obtain. In such situations, it may be of interest to estimate the density of the underlying distribution, or to estimate the reliability of a (series or parallel) system. We investigate such problems in the semiparametric Bayesian framework. In particular, we assume that the observations are from a mixture of exponentials and the parameters of these exponentials have a Dirichlet Process Prior. The effects of prior parameters, the choice of balanced/unbalanced sample, and the number of replications on the estimates are investigated. The method is illustrated by application to water discharge data.
|
- 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 2002 program |
