离散选择模型(Discrete Choice Model,DCM)是一种非线性的多元统计分析方法,主要基于消费者对产品/服务的选择来预测消费者的购买行为。在电信行业的市场营销中,消费者对电信产品的选择行为对于电信运营商的营销策略起着至关重要的作用。通过考虑消费者的特征属性及电信产品(套餐)的特征属性,建立了基于多属性的多项logit(Multinomial Logit,MNL)模型,并通过SPSS软件的MNL回归分析,得到各属性的相关估计参数,最后得到消费者对于每个套餐的选择概率,用来预测消费者选择套餐的行为。另外,针对在校大学生消费群体选择套餐作为案例进行了实证研究,设计了调查问卷并通过在线网络调查及传统问卷调查两种方式获得相关数据,验证了所建MNL模型预测的准确性。
Discrete choice model (DCM) is a kind of statistical method with nonlinear multivariate. In marketing research it can be used to predict consumers'. purchase behavior. Multinomial logit (MNL) model is a kind of DCM, and it has been used in many research areas. In this paper, MNL model was built by considering consumers' characters and packages' features, and MNL regression analysis was conducted via SPSS. The correlative estimation parameters of every attribute are obtained, and the probability of selecting each package is obtained as well. Empirical study of package choosing aiming at college students was performed, and the questionnaire was also used. The relevant data through online survey and paper questionnaire was collected, and the accuracy of the MNL model was verified.