|
Activity Number:
|
540
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Thursday, August 10, 2006 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Statistics and Marketing
|
| Abstract - #306427 |
|
Title:
|
A New Methodology to Modeling Prospects of Corporate Database
|
|
Author(s):
|
Kangkang Xu*+
|
|
Companies:
|
Experian
|
|
Address:
|
955 American Lane, Schaumburg, IL, 60173,
|
|
Keywords:
|
corporate database marketing ; profile model ; panel data
|
|
Abstract:
|
Profile models are built to identify good prospects for clients in huge corporate database. To modeling prospects, there is a common modeling bias in this industry because of corporate database structure. We propose a new modeling design to profile prospects. In stand of using cross sectional data regression, we structure panel data to profile new custom by modeling time series and cross sectional data. We also structure custom purchase behavior by utilizing affinity modeling for purchases that provides a better panel data for modeling. We also make universe pre-selection/clustering to reduce bias that is caused by corporate database structure.
|
- 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 2006 program |
