Publication
Hedging Performance of Multiscale Hedge Ratios
[Paper]
In this study, we combine the wavelet multiscale model and neural network to improve the hedging performance of multiple classes of assets over traditional hedging models.
Decoding Mutual Fund Performance: Dynamic Return Patterns via Deep Learning
[Paper]
2026,
Journal of Financial Stability
This paper uses a sequential deep learning model to uncover and predict dynamic patterns in mutual fund returns. A long-short portfolio based on the model generates a 2.8% annualized four-factor alpha that persists for up to four years.
Working Paper
Can Machines Understand Human Skills? Insights from Analyst Selection
[Paper]
• Revising for invited re-submission at Journal of Finance
We use machine learning to identify skilled financial analysts and aggregate their forecasts into a crowd wisdom-based earnings prediction. Machine-selected analysts persistently outperform expert-picked star analysts by leveraging nonlinear interactions of analyst characteristics rather than relation-based information.
Mapping the Midweek Mountain: The New Geography of Hybrid Work
[Paper]
• Under submission
Using 41 billion mobile geolocation records, we document a lasting post-pandemic shift in work patterns: a 'midweek mountain' of office attendance on Tuesdays through Thursdays, with workers spending significantly more time at non-work locations during the workday.
The Impact of AI Adoption on Hedge Fund Performance
We examine how AI adoption affects hedge fund performance and find that it improves returns by 2.64% annually, reduces risk, and leads to more diversified portfolios with less concentration in local stocks.
The Vulnerability Trap: How Sudden Stops Erode Firm Financial Health Through Supply Chain Destruction
We show that sudden stops in international capital flows destroy firms' supply chain networks, creating a vulnerability trap: financially fragile firms suffer the most network damage, which further weakens their financial health and amplifies exposure to future crises.
Why do actively managed mutual funds hold ETFs? Evidence on liquidity management
This paper examines why actively managed mutual funds hold ETFs and finds they serve as a liquidity management tool. Funds using index ETFs maintain performance regardless of investor flow, while non-users suffer lower returns during redemptions.