On Bias in Social Learning and Consumer Choice

【主讲】陈宁远，香港科技大学工业工程及决策分析学系的助理教授

【主题】社交学习与消费选择的认知偏差

【时间】2019年5月14日（周二）上午10:00am – 12:00pm

【地点】清华经管学院伟伦楼453

【语言】英语

【主办】管理科学与工程系

【简历】Ningyuan Chen的简历

Short Bio

Ningyuan Chen is currently an assistant professor at Department of Industrial Engineering and Decision Analytics in Hong Kong University of Science and Technology.

Dr.Chen’s research interest includes revenue management and dynamic pricing, networks, and statistics. He received his PhD from the Industrial Engineering and Operations Research (IEOR) department at Columbia University, and his BS in Mathematics from Peking University.

【Speaker】Ningyuan Chen, Assistant Professor, Department of Industrial Engineering and Decision Analytics, Hong Kong University of Science and Technology

【Topic】On Bias in Social Learning and Consumer Choice

【Time】Tuesday, May 14, 2019, 10:00am – 12:00pm

【Venue】Room 453, Weilun Building, Tsinghua SEM

【Language】English

【Organizer】Department of Management Science and Engineering

【Abstract】 Reviews for products and services, written by consumers and users, have become an influential input to the purchase decision. For many service businesses they have also become part of the performance review for managers with rewards tied to improvement in the aggregate rating. It is therefore of great importance to understand how much the public ratings reflect true quality of the product or service. Many empirical papers have documented a bias in the aggregate rating arising from various sources --- both consumers' self-selection bias in reporting reviews, as well as potential customers' bounded rationality in evaluating previous reviews. While there is a vast empirical literature, theoretical models that try to isolate and explain the ratings bias are relatively few, and most are based on rational Bayesian learning on the part of consumers. However, writing a review requires some effort (even if firms try to make it as painless as possible) and it seems unlikely -- as well documented and tested in the behavioral economics field -- that consumers make the effort to do a proper Bayesian update of their beliefs before making purchases. In this paper we investigate the nature of the self-selection bias, i.e., consumers confound ex-ante innate preferences for a product or service with ex-post experience and are unable or unwilling to separate the two. We develop a parsimonious dynamic choice model for consumer purchase decisions. We quantify the bias, whose unintended consequence is to make the tastes of consumers look more ``heterogeneous'', benefiting niche products and hurting popular products in terms of the choice probability. We investigate how this affects the firms' assortment and pricing decisions on an online platform, and how this leads to a statistical procedure to estimate the quality variability. The model prediction is validated by empirical evidence using a public IMDb dataset.