----------------------------------------------------------------------------------------------------

               Department of Systems Engineering and Engineering Management

                       The Chinese University of Hong Kong

----------------------------------------------------------------------------------------------------

Date:  Friday, Nov 25, 4:30 pm – 5:30 pm (HKT)

Venue: ZOOM: https://cuhk.zoom.us/j/92598548712?pwd=WmpWenllZGVSZ2g5QzVYc1F2UjZhQT09

Title: A Unified Analysis for Assortment Planning with Marginal Distributions

Speaker: Prof. Selin Damla Ahipasaoglu, School of Mathematical Sciences, University of Southampton, UK

Abstract:

We study assortment problems under the marginal distribution model (MDM), a semiparametric choice model that only requires marginal error information without assuming independence. We characterize the marginal distributions under which a profit-nested assortment is optimal. Moreover, we prove that the best profit-nested assortment is a 1/2-approximate solution for all MDM. These results either generalize or improve existing results on the assortment optimization under the multinomial logit (MNL) model, multiple-discrete-choice model, and the threshold utility model. Lastly, we focus on the marginal exponential model (MEM) as an alternative to capture heteroscedasticity. While assortment problem under MEM remains NP-hard in general, the sufficient conditions for optimality of profit-nested assortments is simpler and a polynomial approximation scheme is also available. Therefore, assortment under MEM has significant computational advantages compared to other choice models that are also designed to capture heteroscedasticity, including Heteroscedastic Extreme Value (HEV) choice model and Heteroscedastic Exponomial Choice (HEC) model. Our numerical studies using synthetic and real-world data sets show that MEM provides competitive predictive and prescriptive performance in capturing heteroscedasticity despite its simple and parsimonious structure.

Biography:

Dr Selin Damla Ahipasaoglu joined the University of Southampton as an Associate Professor in Operational Research within Mathematical Sciences in July 2020. Prior to that, she served as Visiting Faculty member in National University of Singapore, Assistant Professor at Singapore University of Technology and Design and Postdoctoral Researcher at London School of Economics and Princeton University. She received her PhD from the School of Operations Research and Information Engineering at Cornell University in 2009. Her research interests include large scale convex optimisation especially first-order methods, distributionally robust optimisation with marginals, discrete choice models, and optimal experimental design.

Everyone is welcome to attend the talk!

SEEM-5201 Website: http://seminar.se.cuhk.edu.hk

Email: seem5201@se.cuhk.edu.hk