Research Summary

The materials on this page reflect only a small, non-proprietary subset of my research. The majority of my work involves confidential systematic investment research that cannot be shared publicly.

My research focuses on quantitative investment management, integrating economic structure, machine learning, and diverse data sources to generate forecasts for returns, risk, liquidity, and related quantities across horizons ranging from seconds to months. I also work on optimization problems in portfolio construction and trading. This research spans four systematic portfolio areas: equities, futures/macro, event-driven strategies, and their multi-strategy combination. While carrying out this research at Versor, I collaborate with and support a team of more than thirty researchers.

Recent projects include building multi-model machine-learning pipelines for regression and classification forecasts, constructing alpha signals for global futures and FX markets, nowcasting firm fundamentals using alternative data, constructing market-relevant peer groups, decomposing returns into persistent and mean-reverting components, and developing forecasts of merger outcomes. Across strategies, I also develop security-level risk models and transaction-cost estimates that integrate directly with portfolio construction and execution systems. A consistent theme of my work is extracting structure from large, complex datasets and translating it into features and models that support robust forecasting and stable downstream use.

In addition, I produce some non-proprietary work that is listed below.

For up-to-date PDF copies of my publications, please also look at the Versor Investments Athenaeum.