CONSUMER SERARCH AND CHOICE IN ONLINE SHOPPING

This thesis mainly studies the behavior of consumers in searching and choosing products in the process of online shopping. These behaviors are usually related to the features of consumers themselves, the attributes of products and the utility of consumers to the products they expect to buy. This study is based on the relevant data provided by JD.com, including Stock Keeping Unit (SKU) data, user data, consumer click data during browsing and the final purchase order data. This study puts forward and practices three main research directions. The first direction is to cluster products by multidimensional graph clustering method, and segment the user samples that have clicked the products according to the clustering results, and then apply the traditional consumer choice model in each class to obtain the relationship between user attributes and SKU attributes. The second direction is to predict consumers' final purchase decision by combining consumers' attributes, SKU attributes and consumers' click stream. Although the likelihood function of this method is complex, which leads to the difficulty of parameter estimation, this study uses a method similar to Monte Carlo simulation, and verifies the availability of this method to some extent by comparing the simulation results with the real results. The third direction is to cluster the click behavior of user samples by constraining the upper and lower limits of constrained K-means clustering, and then explore the relationship between user click behavior and user attributes through classification algorithm. Lastly, from the managerial perspective, we can get the relationship between user attributes, product attributes and user click process from all directions, so as to guide managers to make business decisions.

[1] Mehta, N., S. Rajiv, and K. Srinivasan (2003). Price uncertainty and consumer search: A structural model of consideration set formation. Marketing science 22 (1), 58–84.
[2] Shen, M., C. S. Tang, D. Wu, R. Yuan, and W. Zhou (2020). Jd. com: Transaction-level data for the 2020 msom data driven research challenge. Manufacturing & Service Operations Management.
[3] Kim, J. B., P. Albuquerque, and B. J. Bronnenberg (2010). Online demand under limited consumer search. Marketing science 29 (6), 1001–1023.
[4] Honka, E. (2014). Quantifying search and switching costs in the us auto insurance industry. The RAND Journal of Economics 45 (4), 847–884.
[5] Honka, E., & Chintagunta, P. (2017). Simultaneous or sequential? search strategies in the us auto insurance industry. Marketing Science, 36(1), 21-42.
[6] Chan, T. Y. and Y.-H. Park (2015). Consumer search activities and the value of ad positions in sponsored search advertising. Marketing Science 34 (4), 606–623.
[7] Kim, J. B., P. Albuquerque, and B. J. Bronnenberg (2017). The probit choice model under sequential search with an application to online retailing. Management Science 63 (11), 3911–3929.
[8] Hortacsu, A. and C. Syverson (2004). Product differentiation, search costs, and competition in the mutual fund industry: A case study of s&p 500 index funds. The Quarterly journal of economics 119 (2), 403–456.
[9] Hong, H. and M. Shum (2006). Using price distributions to estimate search costs. The RAND Journal of Economics 37 (2), 257–275.
[10] De los Santos, B., A. Horta¸csu, and M. R. Wildenbeest (2012). Testing models of consumer search using data on web browsing and purchasing behavior. American economic review 102 (6), 2955–80.
[11] Koulayev, S. (2014). Search for differentiated products: identification and estimation. The RAND Journal of Economics 45 (3), 553–575.
[12] Aouad, A., V. F. Farias, and R. Levi (2015). Assortment optimization under consider-then-choose choice models. Available at SSRN 2618823.
[13] Jagabathula, S., Mitrofanov, D., & Vulcano, G. (2019). Inferring consideration sets from sales transaction data. NYU Stern School of Business.
[14] Tang, W., Lu, Z., & Dhillon, I. S. (2009, December). Clustering with multiple graphs. In 2009 Ninth IEEE International Conference on Data Mining (pp. 1016-1021). IEEE.
[15] Emadi, S. M., Modaresi, S., & Deshpande, V. (2021). Diamonds in the Rough: Leveraging Click Data to Spotlight Underrated Products. Available at SSRN 3835482.
[16] Yamamoto, T. (2011). A multinomial response model for varying choice sets, with application to partially contested multiparty elections. Working paper.
[17] Branco, F., Sun, M., & Villas-Boas, J. M. (2012). Optimal search for product information. Management Science, 58(11), 2037-2056.
[18] Chen, Nan and Clithero, John A. and Hsu, Ming, (2021). Demand Estimation and Forecasting Using Cognitive Models of Consumer Choice.
[19] Croissant, Y. (2012). Estimation of multinomial logit models in R: The mlogit Packages. R package version 0.2-2.
[20] Abdallah, T., & Vulcano, G. (2020). Demand estimation under the multinomial logit model from sales transaction data. Manufacturing & Service Operations Management.
[21] Weitzman, M. L. (1979). Optimal search for the best alternative. Econometrica: Journal of the Econometric Society, 641-654.
[22] Chung, J. H. (2019). Estimation of Sequential Search Models (Doctoral dissertation, The University of Chicago).
[23] Kim, J. B., Albuquerque, P., & Bronnenberg, B. J. (2008). Modeling Online Consumer Product Search.