2021 年 64 巻 1 号 p. 12-30
The purpose of this study is to reveal how marketing affects customers' purchase amount and number of visits in Japanese department stores. We model purchase amounts by using a hierarchical Bayes regression model and number of visits by using a hierarchical Bayes Poisson regression model. Furthermore, we estimate the latent factor behind price as the purchase amount per month with a Type-1 Tobit model and the structural heterogeneity of each customer with a model for variable selection. Direct mail and events are used as marketing measures. The analytical results reveal marketing measures that raise customers' final purchase amounts.