Research Seminar
Constrained latent class models for simultaneous product positioning and market segmentationMichel MeuldersH.U.Brussel and KU Leuven | |
| Abstract: | A key task of strategic marketing is to study the competitive structure of products. This type of analysis is most often based on a spatial configuration or on a categorization of the products. Besides information on the similarity of products, an important goal of competitive structure analysis is to investigate to what extent distinct customer segments prefer specific groups of products, and whether the perception of products depends on customer segments. Candel and Maris (1997) use a probabilistic feature model to categorize products on the basis of latent features. However, in their analysis customer differences are not taken into account. To solve this problem we develop a new probabilistic feature model in which a product positioning and a customer segmentation are derived simultaneously. As an illustration, the model is used to analyze binary judgments of 191 first-year psychology students who indicated for 25 types of sandwich fillings and 14 filling characteristics whether or not a certain sandwich filling has a certain characteristic. |
| Date: | Tue May 4, 12:15 pm - 1:15 pm |
| Place: | room 02.51 (Department of Psychology, Tiensestraat 102, 3000 Leuven) |