Research Summary

Most of my research is proprietary and is not reflected in the publications listed below. 

My research focuses on quantitative investment management, combining machine learning with diverse data sources to generate forecasts for returns, risk, liquidity, and related quantities. These forecasts span time horizons from minutes to months, depending on the application. Additionally, I work on optimization problems in portfolio construction and trading. At Versor, I also lead a research team of 30+ while conducting this work.

For return predictions, we nowcast firms’ fundamentals, compare firms’ and investors’ growth expectations, assess relative valuations, decompose past returns into mean-reverting and persistent components, and analyze sentiment in text and speech. Much of this research extends beyond analysis of financial statements, prices, and volumes.

Across all strategies, we develop proprietary, security-level risk models that account for the risks embedded in our return forecasts. These models cover global equities, merger-related stocks, and futures across all major asset classes.

In parallel, I and my co-authors produce non-proprietary research, shared through publications. This publicly available work, listed below, represents only a small portion of my research and is often intended for a broader, less technical audience.

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