Segmentation of a panel of consumers using clustering of variables around latent directions of preference

A procedure of clustering of variables is discussed and applied for the pur pose of segmenting a panel of consumers. The underlying principle of the me thod is to find K groups of variables (i.e. the consumers) and K latent com ponents such that the consumers in each group are as much correlated as pos sible with the corresponding latent component. The procedure involves runni ng, in a first step, a hierarchical clustering algorithm to determine the a ppropriate number of clusters and an initial partition of consumers. In a s econd step, a partitioning algorithm is carried out in order to improve the solution thus obtained. This clustering approach is illustrated using two real. data sets. On these data sets, the procedure MD-PREF is also performe d and it is shown how it can be complemented by the outcomes of the cluster analysis. In particular, indication about the number of clusters among con sumers is given. (C) 2001 Elsevier Science Ltd. All rights reserved.