Independent analysis of markets, methods and financial judgement.

Research Note
2026

AI in Financial Markets

Machine learning, signal extraction, and risk monitoring in modern financial research.

This research note examines how artificial intelligence is reshaping financial research, from data processing and signal generation to real-time risk monitoring. It focuses on the role of machine learning in identifying non-linear relationships, filtering market noise, improving probabilistic decision-making, and supporting — not replacing — human judgement in investment analysis.

Research Note
2026

Overnight Price Jumps and Short-Term Return Predictability

Market overreaction, price discovery, and trading implications in U.S. equities.

Overnight price jumps are rare but economically important discontinuities between the previous close and the next open. Large-sample evidence on 9,283 U.S. stocks from 1993 to 2021 shows that both positive and negative overnight jumps predict short-run returns with the opposite sign, consistent with temporary overreaction followed by correction once continuous trading resumes. The reversal is strongest immediately after the open and decays quickly as the holding horizon extends. The note also discusses what the move toward 23-hour U.S. equity trading may mean for this effect.