
简介
理悦CEMA•玉衡系列讲座第一讲将于4月7日(周二)下午13:30-15:00在学术会堂712会议室举行,由由湖南大学经济与贸易学院副教授杨湘黔老师报告金融理论论文“Optimal information in sequential learning”(序贯学习中的最优信息)欢迎广大师生(尤其是对科研有兴趣和追求的本科生、硕士生和博士生)参加。讲座对所有人免费开放,不需要提前预约。
题目
Optimal information in sequential learning
(序贯学习中的最优信息)
报告人
杨湘黔
湖南大学经济与贸易学院副教授,本科毕业于中央财经大学中国经济与管理研究院(优秀校友),博士毕业于新加坡国立大学经济系,主要研究方向为信息设计与机制设计,目前主持国家自然科学基金青年项目1项。代表作发表在国际一流的Journal of Economic Theory和International Economic Review,并有文章在博弈论一流期刊Games and Economic Behavior返修中。
摘要
We study information design in a differentiated-product duopoly engaging in sequential pricing competition. Using a dynamic indifference approach, we demonstrate that any pure-strategy subgame perfect equilibrium (SPNE) outcome can be implemented by a canonical information structure with a hybrid geometry. Contrary to the unit-elastic constructions that characterize optimal information under monopoly or simultaneous-move price competition, our canonical information structure is a family of distributions transitioning from a Lambert W function—capturing the leader’s trade-off between marginal price gains and the follower’s endogenous best response—to the unit-elastic hyperbola that binds the follower’s iso-profit constraint. The canonical information structure is parametrized by only three variables, allowing us to derive the optimal information for the leader, the follower, and the consumer, respectively. The leader-optimal information induces a strong (negative) price gap to relax competition. In contrast, null information with a zero price gap is follower-optimal and allows the follower to extract the full surplus of their own product. Finally, consumer-optimal information induces a moderate negative price gap to balance price rivalry and allocative efficiency. Our results sharply contrast with the findings of Armstrong and Zhou (2022), where optimal information consistently induces a zero price gap under simultaneous-move competition.
本文研究差异化产品双寡头在序贯定价竞争中的信息设计问题,运用动态无差异方法证明,任何纯策略子博弈完美均衡结果均可由具备混合几何特征的典范信息结构实现。与垄断或同时定价竞争下最优信息所呈现的单位弹性构造不同,本文的典范信息结构为一族分布形式,其形态从反映领先者在边际价格收益与跟随者内生最优反应间权衡的朗伯W函数,逐步过渡至约束跟随者等利润约束的单位弹性双曲线,该结构仅由三个变量参数化,据此可分别推导得出领先者、跟随者与消费者各自最优的信息结构。领先者最优信息会引致显著为负的价格差距以缓和市场竞争,而价格差距为零的空信息则是跟随者的最优选择,可使其抽取自身产品的全部剩余,消费者最优信息则会形成适度的负向价格差距,以平衡价格竞争与配置效率。本文结论与Armstrong和Zhou(2022)的研究形成鲜明对比,在其同时行动竞争框架下,最优信息始终会引致零价格差距。
时间
4月7日13:30-15:00
地点
学术会堂712
活动对象
创新发展学院师生
主办单位
创新发展学院
中国经济与管理研究院
撰稿:杨湘黔
审稿:何其春
编辑:沈嘉怡
审核:赵扶扬