The scope of this repository is to keep track of state of the art (SoTA) results for machine learning (ML) algorithms applied in quantitative trading domain. Specifically, this repo is to collect and report quantitative trading model performance metrics under diffenent types of environments, including
- Academic/Industrial
- Train/Dev/Test dataset
- Backtest/Paper-trading/Production
this is an attempt to release
- a set of concrete and reliable benchmarks
- state-of-the art(SoTA) results
to help reseacher explore and expand the boundary of predictability.
We're looking for pull requests related to results from papers or reliable disclosure we should add, and better organization of the results.
| Title | Author | Algorithm | frequency | date_range | adj.R2 | mse | year |
|---|---|---|---|---|---|---|---|
| On the predictability of Chinese stock returns | Xuanjuan Chen and Tong Yu | regression | monthly | 1995.06 - 2007.07 | 9% | Na | 2009 |
| Short- and Long-Horizon Behavioral Factors | Daniel and Lin Sun | regression | monthly | 1972.07 - 2014.12 | 7.6% | Na | 2018 |
| Title | Author | Algorithm | frequency | date_range | predicted R2 | mse | year |
|---|---|---|---|---|---|---|---|
| Empirical Asset Pricing via Machine Learning | Shihao Gu,Bryan T. Kelly, and Dacheng Xiu | comprehensive | monthly | 1957.03 - 2016.12 | 0.39% per month | Na | 2018 |
Fama-Miller Center: for Research in Finance
Applied Quantitative Research(AQR):AQR is a pioneer as a quantitative investor and as a publisher of influential academic research.
Oxford-Man Institute of Quantitative Finance:We answer fundamental questions about financial markets, and develop new quantitative methods and insights with the potential to transform them.