Postdoctoral Researcher
Operations, Technology, and Information Management (OTIM)
Cornell SC Johnson College of Business,
Research Interests:
Context : Online Marketplaces and Retail Operations
Topics : Online Grocery Purchases, Ride-Hailing Choices, Demand Estimation
Methods : Structural Estimation, Machine Learning, Generative AI
I am a postdoctoral researcher at the Operations, Technology, and Information Management (OTIM) department at the Cornell SC Johnson College of Business, Cornell University. I completed my Ph.D. in the Operations Management department at NYU Stern School of Business, advised by Professor Srikanth Jagabathula. My research focuses on using ML and GenAI methods to predict customer purchase choices in online marketplaces to help firms enhance their operational strategies. I accurately model customer decision-making processes by integrating often-overlooked factors such as sequential decision-making, consideration sets, and behavioral characteristics into structural choice models. I collaborate with leading companies to acquire large-scale, real-world data encompassing millions of data points from hundreds of thousands of customers. Utilizing state-of-the-art ML techniques, including Bayesian estimation and Generative AI, I effectively estimate these structural models, providing firms with optimal operational strategies for promotions, pricing, and product assortments.
My research portfolio includes three papers focused on enhancing promotional and other operational strategies by analyzing critical customer decisions at three stages of a purchasing process: (1) choosing a platform, (2) selecting categories within a platform, and (3) selecting a product within a category.
This paper examines platform choice by analyzing how customers select ride-hailing services like Uber and Lyft to enhance promotional strategies. We evaluate the impact of operational factors, such as price and wait time, as well as behavioral factors, including platform stickiness and search behavior.
This paper investigates how customers sequentially choose product categories to add to their shopping carts and differentiate between items bought due to inherent customer demand and those purchased primarily to take advantage of cart-level promotions, such as free delivery thresholds.
This paper examines product choice within a single category to estimate product demand—a crucial input for operational decisions like pricing and assortment optimization—while accounting for unobserved factors such as product shelf placement, brand awareness, and local events.