Discrete Choice Model of Customer Behavior and Empirical Study

摘要

Discrete choice model is an important theoretical basis for the study of customer behavior. Relative to the classic multinomial logit (MNL) and nested logit (NL) model, due to the new marginal distribution model (MDM) and marginal moment model (MMM) model, the limiting condition is very weak and the applicability is strong, and they can be widely used in the development of product line. This paper studies the effectiveness of MDM and MMM. Firstly, four models and their characteristics are discussed, and the influence factors of customer purchase behavior in different models are analyzed by empirical research and multiple nonlinear regression. Theoretical analysis and empirical results show that, with the diversification and the personalization of customer demand, MDM and MMM model are more suitable for the actual results, and the results are better than MNL and NL model, and they have broad application prospects. At the same time, this paper will have some positive effect on the design of enterprise marketing plan.