My recent research is focused on applications of machine learning (ML) and AI in asset pricing.

We are quickly realizing that applying big data and AI/ML to such standard problems like asset allocation, optimal portfolio construction, risk management and investments overall provides significant incremental value compared to using traditional tools.

Yet off-shelf AI algorithms are not well suited for low signal-to-noise ratio of financial data. In my research, with my co-authors, we show how incorporating economic assumptions and restrictions into AI algorithms helps improving investment decision making processes.

It can. Using textual information from a complete history of regular quarterly and annual filings by U.S. corporations, we train classic machine learning algorithms and large language models, LLMs, to predict future earnings surprises. Only finance-objectives trained LLMs have the capacity to “understand” the contexts of previous 10-Q (10-K) releases to predict both positive and negative earnings surprises, and future firm returns.