|
Activity Number:
|
91
|
|
Type:
|
Invited
|
|
Date/Time:
|
Monday, August 7, 2006 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Physical and Engineering Sciences
|
| Abstract - #305038 |
|
Title:
|
Forecasting Warranty Claims
|
|
Author(s):
|
Jerald F. Lawless and Marc Fredette*+
|
|
Companies:
|
HEC Montréal and University of Waterloo
|
|
Address:
|
3000 Chemin de la Cote Sainte Catherine, Montreal, PQ, H3T 2A7, Canada
|
|
Keywords:
|
prediction ; repeated events ; Poisson processes ; random effects
|
|
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. Prediction intervals are sought. These will initially be based on limited data concerning the units in question, and thus be rather wide. As warranty claims data accumulate, the intervals will narrow. Methods based on Poisson processes with random effects, combined with calibration by simulation, will be used to generate frequentist prediction intervals. An application to the prediction of automobile warranty claims will be considered.
|
