A Generalized Ordinal Finite Mixture Regression Model for Market Segmentation Yifan Zhang, Duncan K. H. Fong, and Wayne S. DeSarbo
International Journal of Research in Marketing Vol. 38, No. 4 (December 2021), pp. 1055-1072
Overview The importance of market segmentation is widely acknowledged. It can guide a firm’s marketing strategy and resource allocation among products and markets. With recent advancements in information technologies, many firms now have enhanced capabilities to capture data about their customers and business operations. In this article, we show that having more data on customers does not always improve market segmentation analysis; in fact, the inclusion of unimportant variables and erroneous observations can distort and mislead. To solve the problem, we develop a new procedure that simultaneously performs segmentation and variable selection within each derived segment. The procedure is robust to data errors and can include concomitant variables for simultaneous profiling of the derived segments. Using both synthetic and empirical data, our model outperforms other segmentation tools in parameter recovery, segment retention, and segment-membership prediction.
16 | Journal Publications - Distinguished Journal
Made with FlippingBook - Online catalogs