Abstract
Differences in preference models between segments often are influenced by the background characteristics of individuals. Based on the relationship between preference weights and background characteristics, predictions of preference are made for new segments and geographic areas. Preference models are found to be transferable in some cases, with better predictions for new segments than for geographic areas. Generalizability of results, limitations of research, and conditions for finding transferable models are discussed.
Get full access to this article
View all access options for this article.
