Activity Number:
|
313
|
Type:
|
Contributed
|
Date/Time:
|
Wednesday, August 14, 2002 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Section on Physical & Engineering Sciences*
|
Abstract - #301275 |
Title:
|
Estimation and Inference for Failure Rate in the Presence of Incomplete Failure Data
|
Author(s):
|
Jeffrey Glosup*+
|
Affiliation(s):
|
Sun Microsystems
|
Address:
|
901 San Antonio Road, Mail Stop UMPK18-216, Palo Alto, California, 94303, USA
|
Keywords:
|
reliability data ; EM algorithm ; exponential distribution
|
Abstract:
|
Estimation and inference for the failure rate of an exponential distribution is well understood in situations where the exact number of failures is known. However, in the collection of field data, recorded failures may include multiple components or causes. In some cases, data indicating the component failing or cause of failure may be missing, leading to incomplete data on the number of failures. In this presentation, methods based on the EM algorithm are developed for using this incomplete data in the estimation of the failure rate for an exponential distribution. Methods for use of incomplete data in statistical inferences involving the failure rate are also explored.
|
- 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 |