|
Activity Number:
|
507
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Thursday, August 2, 2007 : 8:30 AM to 10:30 AM
|
|
Sponsor:
|
Business and Economics Statistics Section
|
| Abstract - #309467 |
|
Title:
|
Evaluating Forecasts
|
|
Author(s):
|
Wanli Min*+ and Chaitra Nagaraja
|
|
Companies:
|
IBM T.J. Watson Research Center and University of Pennsylvania
|
|
Address:
|
1101 Kitchawan Road, Route 134, Yorktown Heights, NY, 10598,
|
|
Keywords:
|
forecasting ; model averaging ; boosting
|
|
Abstract:
|
Forecasting problems arise in every aspect of business from predicting stock prices to product revenue. Innumerable approaches can be taken to model any specific problem. As business environments are continuously changing, there is often no paramount model. In such a case, it is valuable to generate several models and average over them. While this increases the stability, and hopefully accuracy, of the estimates, it also increases the complexity of the model. The latter feature makes evaluating the confidence of a forecast difficult. We propose a method to solve this issue using boosting procedures.
|
- 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 2007 program |
