A novel statistical approach to marketing campaigns

We propose a novel heterogeneous choice model to deal with heterogeneity issues in decision making across individuals for marketing campaigns. Based on the proposed model, we further develop a novel classification method that enables us to identify an optimal group of individuals in a database for, e.g. future bank marketing campaigns. Our results show that, in the presence of heterogeneity, our model outperforms some commonly used benchmark models with respect to model specification errors, and our classification method outperforms the conventional method in respect of the positive response rate in marketing campaigns.