- Identify Brand Satiation Using Purchase Data (With Venkatesh Shankar) (recent version)
In product categories such as yogurt, cereal and candy, consumers are likely to be satiated after frequent consumption of the same brand, leading to variety-seeking and switching to other brands. Prior research has modeled satiation mostly using consumption and preference data, but most firms have access to only purchase data. Identifying satiation and estimating satiation effect using purchase data remain a significant challenge. We provide rich evidence supporting effects of satiation using a scanner data set in a yogurt market and develop a Hidden Markov Model in which households may be temporarily stay in an unobserved “satiation” state. The results show that households may be occasionally satiated for a certain brand, and there is significant difference among the satiation probabilities for brands. Our Hidden Markov Model explains consumer satiation better than benchmark models.
- The Effect of Periodic Structure in Consumer Visiting Patterns (With Hua Yuan) (recent version) (original working paper)
Maintaining strong periodic shopping trips may be not easy for households, because they face uncertainties from consumption, external shock and other schedule conflicts. While the routinization of production makes households more efficient, the routinzation of shopping trips brings little benefit. What are the implications for those households who remain strong revisit patterns? We propose a distribution-wise measure of households’ revisit periodicity strength, and investigate its impacts on product choices using scanner datasets. The product-market level analysis shows that households with strong periodic revisit patterns are associated with weaker consumer inertia and have more product switches recorded. The data is consistent with the explanation that those households seek compensatory variety in product choice due to unobserved constraints in timing choice.
- Explain Heterogeneity in State Dependence using “Fundamental” Switches (recent version)
This paper investigates state dependence effects in frequently purchased product markets. I use consumers’ switching behavior in different product categories to provide necessary variation and test whether the variation may explain differences in consumers’ responses to previous purchases and other relevant marketing variables. I find that part of the variation in switching behavior is stable: they explain a significant portion of consumers’ state dependence in different categories and different years consistently. The finding provides a different way to look at household switches, and contributes to the literature of estimating state dependence in the consumer goods market.
- Customers’ shopping behavior on the E-commerce Platform (with Lengyang Wang)
Utilizing a transaction-based dataset, we hope to investigate some purchase patterns on E-commerce platforms (which cannot be studied previously). Our mission is to think of a reasonable way that customers make choices on the platform. With a reasonable choice structure, we may address problems such as comparing gender differences in online purchasing behavior.