Pure Characteristics Demand Models under Stochastic and Distributionally Robust Environment

報告題目：Pure Characteristics Demand Models under Stochastic and Distributionally Robust Environment

報 告 人：孫海琳 教授  南京師范大學(xué) 優(yōu)青

邀請人：趙志華  

報告時間：2025年4月25日（周五） 11:00-13:15

報告地點：北校區(qū)會議中心303-1

報告人簡介：孫海琳，南京師范大學(xué)數(shù)學(xué)科學(xué)學(xué)院教授，副院長，國家級青年人才。孫教授于2007年在吉林大學(xué)獲得統(tǒng)計學(xué)學(xué)士學(xué)位，2013年畢業(yè)于哈爾濱工業(yè)大學(xué)，獲數(shù)學(xué)博士學(xué)位。在其博士期間，先后前往英國南安普頓大學(xué)和香港理工大學(xué)聯(lián)合培養(yǎng)。2015-2017年在香港理工大學(xué)應(yīng)用數(shù)學(xué)系做博士后研究。2018年獲中國運籌學(xué)會青年科技獎和江蘇省數(shù)學(xué)成就獎，主持國家自然科學(xué)基金優(yōu)秀青年科學(xué)基金項目、面上項目和青年科學(xué)基金項目。他的研究領(lǐng)域包括隨機(jī)優(yōu)化，分布魯棒優(yōu)化、隨機(jī)變分不等式及其在投資組合、風(fēng)險管理和經(jīng)濟(jì)學(xué)模型上的應(yīng)用。他在包括《Mathematical Programming》《SIAM Journal on Optimization》《Mathematics of Operations Research》等國際權(quán)威期刊發(fā)表了二十多篇論文。

報告摘要：Pure characteristics demand models is a utility-based choice model which is used to determine a consumer’s purchase decision among a list of available products and to provide an estimate of product demands. Under stochastic and distributionally robust environment, we consider the two challenge problems: the problem of estimating model parameters based on generalized method of moments and the multiproduct pricing problem based on pure characteristics demand models. In stochastic case, the regularized sample average approximation method is used to handle the set-valued stochastic equilibrium constraints. In distributionally robust case, we use the regularization method and the data driven approach to handle the set-valued distributionally robust equilibrium constraints. The corresponding convergence analysis results are given and numerical results are reported to show that validate the effectiveness and scalability of our approach.

主辦單位：數(shù)學(xué)與統(tǒng)計學(xué)院