My research lies at the intersection of revenue management, pricing, and matching markets, with a focus on how firms can design pricing and allocation strategies in environments shaped by uncertainty, heterogeneity, and large-scale data. Across my projects, I develop theory-driven models that provide both structural insights and implementable solutions, aiming to bridge rigor and managerial relevance.
This paper studies the optimal dynamic pricing of split-stay products, where a trip is divided between two accommodations. Using a dynamic programming framework with linear random utility models, I derive the structure of optimal pricing policies and show that split stays play an unexpected inventory-balancing role. They are used primarily in the mid-range of the selling horizon, when the need to rebalance capacity is most acute. The results highlight that split stays are not only a product innovation but also a powerful lever for revenue and inventory management.
Major Revision at Management Science
๐ [Download Paper]
Presale vouchers are widely sold by hotels at a discount, are fully refundable, and are redeemed later subject to room availability. Because presale revenue is earned only when a redemption is fulfilled, and market buyers compete for the same capacity, the hotel must jointly decide its market prices, its voucher price, and how much capacity to release to redemption on each night. This paper studies this problem in a model that couples price optimization with online stochastic matching on shared capacity, and characterize the fulfillment guarantee the seller can achieve under uncertain redemption preferences.
Submission scheduled October 2026.
๐ PDF available upon request
This paper develops a linear programming framework for frictional matching markets with heterogeneous agents and additive search costs. The formulation connects the matching literature to classical optimal assignment problems, allowing us to prove equilibrium properties such as positive assortative matching. Numerical experiments explore how search frictions affect social welfare and suggest policy tools, such as unemployment benefits, to mitigate inefficiencies.
๐ PDF available upon request
This study integrates descriptive, predictive, and prescriptive analytics to evaluate the operational effectiveness of AI-assisted reminder systems in reducing 30-day hospital readmissions among patients with diabetes. Rather than treating prediction as an end in itself, we embed risk stratification within a structured decision framework that explicitly accounts for limited outreach capacity and real-world discharge workflows. Applying this framework to a large publicly available multi-hospital dataset, we find that targeting the top 10 to 20 percent of risk-ranked patients captures 25 to 40 percent of the total achievable reduction in 30-day readmissions while requiring substantially fewer interventions than universal outreach. Beyond this threshold, additional coverage yields diminishing marginal returns.
Health Management, Policy and Innovation, Volume 11, Issue 2 (2026). Published.
๐ [Download Paper]
LIBOR Transition into SOFR (2019โ2020) โ Modeled interest rate term structures using Vasicek and Hull-White models and applied ARIMA/GARCH methods for time series analysis. Selected as a representative project at the 2020 SOFR Summit.
Undergraduate Thesis: Extensions of the Black-Scholes Model (2018โ2019) โ Developed and simulated option pricing models including Black-Scholes, Heston, and GARCH, applying Monte Carlo methods for equity derivative pricing. Completed as senior thesis at the University of Notre Dame.