[1L3-GS-10] AI application: Behavioral economics

[1L3-GS-10-01] Behavioral Economic Forecasting of Store-Visit Behavior Using Location Information Data

〇Ryoya Nakano¹, Shinnosuke Hayama¹, Yuto Murata¹, Rin Tsujikawa¹, Takahiro Hoshino^1,2 (1. Keio University, 2. RIKEN Center for Advanced Intelligence Project)

Keywords:machine learning, location information data, behavioral economics, retail store, store choice

Retail companies are increasingly using databases to predict customer visits, customize sales promotions, and plan store opening strategies. However, current studies mainly rely on purchase data from their own or related stores, overlooking individual behavioral histories and visits to other stores. To bridge this gap, our study introduces a novel store visit forecasting model based on individual-level location data, from the perspective of behavioral economics. Our model expands research opportunities as it solely requires location and phone book data: By integrating variables such as habit formation, loyalty, and exploratory behavior, we improve the accuracy of store visit forecasts without using purchase data from each retail store. Our analysis shows the effectiveness of these behavioral economic variables, underscoring their significant role in the predictive model. This research advances the understanding of customer behavior and provides practical insights for retailers seeking to optimize marketing strategies and store operations.