John Moody
John E. Moody is a researcher in computational finance and machine learning, associated in the relevant period with the Computational Finance Program at the Oregon Graduate Institute of Science and Technology (Beaverton, Oregon) and later affiliated with ICSI Berkeley. He is the lead originator of the direct/recurrent reinforcement learning approach to trading: with Lizhong Wu he published “Optimization of Trading Systems and Portfolios” (1997), and with Wu, Yuansong Liao and Matthew Saffell the Journal of Forecasting paper Moody Wu Liao Saffell 1998, which introduced recurrent reinforcement learning (RRL) for trading systems and portfolios and the Differential Sharpe Ratio as an online risk-adjusted reward.
His best-known work is Moody and Saffell 2001, “Learning to Trade via Direct Reinforcement” (IEEE Transactions on Neural Networks), co-authored with Matthew Saffell — the most-cited reference in the RL-trading literature and the conceptual root of the modern Reinforcement Learning Trading Policy family. He appears in this vault as the founding figure of Recurrent Reinforcement Learning Trading: the idea of skipping a price-forecasting step and instead training a recurrent policy by gradient ascent to maximise a financial performance function net of transaction costs.
John Moody [defines] Recurrent Reinforcement Learning Trading John Moody [defines] Differential Sharpe Ratio
The vault treats the foundational status as confirmed, but notes — consistently with the skeptical research goal — that the empirical profitability claims in Moody’s 1998-2001 papers are dated single-group backtests on data ending in 1994/1996, not independently replicated, and that the strongest modern direct-RRL FX study (Borrageiro Firoozye Barucca 2022) reports far weaker net results.
Connections
- Moody and Saffell 2001 — proposes_model, 2001, source: https://ieeexplore.ieee.org/document/935097/
- Moody Wu Liao Saffell 1998 — proposes_model, 1998, source: https://link.springer.com/chapter/10.1007/978-1-4615-5625-1_10
- Matthew Saffell — relates (long-running co-author), source: https://pubmed.ncbi.nlm.nih.gov/18249919/
- Recurrent Reinforcement Learning Trading — proposes_model, source: https://ieeexplore.ieee.org/document/935097/
- Differential Sharpe Ratio — proposes_model, source: https://proceedings.neurips.cc/paper_files/paper/1998/hash/4e6cd95227cb0c280e99a195be5f6615-Abstract.html
- Reinforcement Learning Trading Policy — proposes_model, source: https://ieeexplore.ieee.org/document/935097/