HETEROGENEOUS DEMAND EFFECTS OF RECOMMENDATION STRATEGIES IN A MOBILE APPLICATION: EVIDENCE FROME CONOMETRIC MODELS AND MACHINE-LEARNING INSTRUMENTS

• 成果类型: Article
• 署名作者: Adamopoulos, Panagiotis; Ghose, Anindya; Tuzhilin, Alexander
• 署名单位: Emory University; New York University
• 刊物名称: MIS QUARTERLY
• ISSN/ISSBN: 0276-7783
• DOI: 10.25300/MISQ/2021/15611
• 发表日期: 2022
• 页码: 101-150
• 关键词: product diversity; empirical-analysis; search costs; long tail; systems; CHOICE; IMPACT; performance; generation; agents

摘要

In this paper, we examine the effectiveness of various recommendation strategies in the mobile channel and their impact on consumers' utility and demand levels for individual products. We find significant differences ineffectiveness among various recommendation strategies. Interestingly, recommendation strategies that directly embed social proofs for the recommended alternatives outperform other recommendations. In addition, recommendation strategies combining social proofs with higher levels of inducedawareness due tothe prescribedtemporal diversity have anevenstronger effect on the mobile channel. We alsoexamine theheterogeneity of the demand effect across items, users, and contextual settings, further verifying empiricallythe aforementionedinformationandpersuasionmechanisms andgeneratingrichinsights. We alsofacilitatethe estimationof causal effects in the presence of endogeneity usingmachine-learning methods. Specifically, we developnovel econometric instruments that capture product differentiation(isolation)basedondeep-learning models ofuser-generated reviews. Our empirical findings extend the current knowledge regarding the heterogeneous impact of recommender systems, reconcile contradictory prior results in the related literature, and have significant business implications