Modeling Preferences for Common Attributes in Multicategory Brand Choice
提出一个多品类品牌选择模型，认为品牌内在效用取决于跨品类共有的属性，家庭对这些属性的偏好存在相关性，模型通过降维捕捉这些共同偏好，并应用于零食和食品品类数据，发现品牌名和低脂属性偏好高度相关。
The authors propose a multicategory brand choice model based on the conceptualization that the intrinsic utility for a brand is a function of underlying attributes, some of which are common across categories. The premise is that household preferences for attributes that are common across categories are likely to be correlated. The model that the authors develop projects the unobserved preferences for attributes to a lower dimensional space of unobserved factors. The factors are interpretable as household “traits” that transcend categories, and they can be used to predict preferences for attributes in new categories. The authors apply the proposed model to household panel data for three closely related snack categories and for two less-related food categories. The authors find strong correlations in preferences for product attributes such as brand names and low fat or fat free. This study demonstrates that these high correlations in product attribute preferences across categories are useful in targeting activities in existing and new categories.