Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
5106329 | International Journal of Forecasting | 2017 | 9 Pages |
Abstract
Marketing applications often require disaggregate forecasts of demand that pertain to subsets of individuals who are targeted for action. Examples include targeted price promotions that are made available through on-site couponing and forecasts of market segments for which new products have been developed. One challenge in the production of disaggregate forecasts of demand, and of consumer responses to marketing actions, relates to the limited amount of data that is available at the individual level. This paper discusses approaches to the improvement of marketing forecasts through the use of both parsimonious structural models of demand and random-effect models that pool data statistically across individual consumers.
Keywords
Related Topics
Social Sciences and Humanities
Business, Management and Accounting
Business and International Management
Authors
Greg M. Allenby,