5 papers in the archive (2024–2026),
average empirical rigor 6.6/10, average math complexity 4.8/10, 1 with linked code. Quadrant mix: Holy Grail ×2, Street Traders ×3.
Scores are our LLM-assisted heuristics (methodology), not peer review or citation counts. Author pages are generated from paper metadata; name variants may split one person across pages.
arXiv author page →
Papers, ranked by score
Ordered by a blend of empirical rigor (60%) and math complexity (40%).
Financial research agents can generate and monitor trading rules faster than a human can document the resulting selection process. Candidates appear at data-dependent times, weak rules are abandoned, and promising ones are watched until they cross a reporting threshold. We introduce PROMOTE, a…
Holy GrailMath 7.5Rigor 8.0
Costs: ModeledOut-of-sample: HoldoutResolution: Daily Data
More details
Authors
Eric Benhamou, Jean-Jacques Ohana, Beatrice Guez, Ethan Setrouk, Chamyl Saadi, Thomas Jacquot
Different learning systems can make nearly the same trade when they share data, objectives, or representations. The market then sees decision concentration even if the models look diverse. We call exposure to this shared direction a signal collision. A linear-impact game shows the basic…
Holy GrailMath 7.5Rigor 6.0capped
Costs: ModeledOut-of-sample: UnclearResolution: Weekly-Monthly Data
More details
Authors
Eric Benhamou, Jean-Jacques Ohana, Beatrice Guez, Ethan Setrouk, Chamyl Saadi, Thomas Jacquot
In this paper, we demonstrate that non-generative, small-sized models such as FinBERT and FinDRoBERTa, when fine-tuned, can outperform GPT-3.5 and GPT-4 models in zero-shot learning settings in sentiment analysis for financial news. These fine-tuned models show comparable results to GPT-3.5 when it…
This paper introduces a new risk-on risk-off strategy for the stock market, which combines a financial stress indicator with a sentiment analysis done by ChatGPT reading and interpreting Bloomberg daily market summaries. Forecasts of market stress derived from volatility and credit spreads are…
We used a dataset of daily Bloomberg Financial Market Summaries from 2010 to 2023, reposted on large financial media, to determine how global news headlines may affect stock market movements using ChatGPT and a two-stage prompt approach. We document a statistically significant positive correlation…