Focal MNL模型:消费者选择的阈值效应,选品,定价与估计问题研究

讲座通知

图片

2026 年 5 月 13 日(星期三),

上午 10:00 - 11:30

图片

信息管理与工程学院102室
上海财经大学(武东路校区)

上海市杨浦区武东路100号

主题

The Focal Multinomial Logit Model: Threshold Effects on Consumer Choice, Assortment, Pricing and Estimation.

图片

主讲人

聂嘉明

深圳大学

聂嘉明教授于2017年在清华大学数学科学系取得运筹学博士学位。于2017年-2020年,在中山大学(珠海校区)数学学院担任副研究员,于2020年-2024年,在厦门大学数学科学学院担任助理教授,于2024年起担任深圳大学管理学院副教授。聂老师的研究方向十分广泛,包括组合优化,近似算法与算法复杂度,以及收益管理等等。其多篇代表工作已发表在管理学顶级期刊(如M&SOM,POMS,IJOC等)以及计算经济学顶级会议(如ACM EC等)上。

讲座简介

In this talk, we consider operations management problems under a focal multinomial logit (FMNL) model, which generalizes the MNL framework and captures irrational choice behaviors—such as context, halo, and choice overload—via focality. We focus on several threshold-based specifications, including constant, cardinality, and linear threshold FMNL models, as well as a broader regular class. We analyze computational complexity and develop polynomial-time exact or approximation algorithms for assortment optimization under different focal structures, as well as for optimal joint pricing and assortment decisions. We also study the maximum-likelihood estimation (MLE) problem, addressing identifiability, consistency, and convergence rates, and we propose a convex MINLP reformulation for global parameter estimation. This is joint work with Lei Guan (Beijing Institute of Technology), Ruxian Wang (Johns Hopkins Carey Business School), and Lianmin Zhang (CUHK, Shenzhen).