Book chapter
FORECASTING HOUSEHOLD RESPONSE IN DATABASE MARKETING: A LATENT TRAIT APPROACH
Advances in Business and Management Forecasting, pp.109-131
Advances in Business and Management Forecasting, Elsevier
01/01/2009
DOI: 10.1108/S1477-4070(2009)0000006008
Abstract
Database marketers often select households for individual marketing contacts using information on past purchase behavior. One of the most common methods, known as RFM variables approach, ranks households according to three criteria: the recency of the latest purchase event, the long-run frequency of purchases, and the cumulative dollar expenditure. We argue that RFM variables approach is an indirect measure of the latent purchase propensity of the customer. In addition, the use of RFM information in targeting households creates major statistical problems (selection bias and RFM endogeneity) that complicate the calibration of forecasting models. Using a latent trait approach to capture a household's propensity to purchase a product, we construct a methodology that not only measures directly the latent propensity value of the customer, but also avoids the statistical limitations of the RFM variables approach. The result is a general household response forecasting and scoring approach that can be used on any database of customer transactions. We apply our methodology to a database from a charitable organization and show that the forecasting accuracy of the new methodology improves upon the traditional RFM variables approach.
Details
- Title: Subtitle
- FORECASTING HOUSEHOLD RESPONSE IN DATABASE MARKETING: A LATENT TRAIT APPROACH
- Creators
- Eddie Rhee - Stonehill CollegeGary J. Russell - University of Iowa
- Contributors
- K D Lawrence (Editor)R K Klimberg (Editor)
- Resource Type
- Book chapter
- Publication Details
- Advances in Business and Management Forecasting, pp.109-131
- Publisher
- Elsevier; KIDLINGTON
- Series
- Advances in Business and Management Forecasting
- DOI
- 10.1108/S1477-4070(2009)0000006008
- ISSN
- 1477-4070
- Number of pages
- 23
- Language
- English
- Date published
- 01/01/2009
- Academic Unit
- Marketing
- Record Identifier
- 9984380405702771
Metrics
3 Record Views