|
Activity Number:
|
115
|
|
Type:
|
Invited
|
|
Date/Time:
|
Monday, August 4, 2008 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Physical and Engineering Sciences
|
| Abstract - #300147 |
|
Title:
|
Using Experts' Knowledge To Forecast Warranty Claims
|
|
Author(s):
|
Marc Fredette*+
|
|
Companies:
|
HEC Montreal
|
|
Address:
|
3000 Côte-Sainte-Catherine, Montreal, QC, h3t 2a7, Canada
|
|
Keywords:
|
Warranty Claims ; Prediction ; Random Effects Models ; Poisson Processes ; Prior Information
|
|
Abstract:
|
We discuss methods for predicting the eventual total number of warranty claims in a large population of units manufactured and sold over a period of time. As warranty claims data accumulate, methods based on Poisson processes with random effects often give appropriate frequentist prediction intervals. However, these methods fail to provide adequate intervals when they are based on early or limited data. Models' extensions incorporating experts' knowledge and other types of prior information will be discussed, and an application to the prediction of automobile warranty claims will be considered.
|
- 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 2008 program |
