Papers, ranked by score

Ordered by a blend of empirical rigor (60%) and math complexity (40%).

Recurrent Neural Networks with more flexible memory: better predictions than rough volatility

Challet et al.

We extend recurrent neural networks to include several flexible timescales for each dimension of their output, which mechanically improves their abilities to account for processes with long memory or with highly disparate time scales. We compare the ability of vanilla and extended long short term…

Holy Grail Math 7.5 Rigor 8.0 Code
Costs: Not applicableOut-of-sample: HoldoutResolution: Daily Data
More details
Authors
Damien Challet, Vincent Ragel
Evaluation
Forecast accuracy

Interpretable ML for High-Frequency Execution

Fabre et al.

Order placement tactics play a crucial role in high-frequency trading algorithms and their design is based on understanding the dynamics of the order book. Using high quality high-frequency data and a set of microstructural features, we exhibit strong state dependence properties of the fill…

Holy Grail Math 6.5 Rigor 8.0
Costs: ModeledOut-of-sample: UnclearResolution: L2 Order Book
More details
Authors
Timothée Fabre, Vincent Ragel
Evaluation
Unclear

Consistent time travel for realistic interactions with historical data: reinforcement learning for market making

Ragel et al.

Reinforcement learning works best when the impact of the agent’s actions on its environment can be perfectly simulated or fully appraised from available data. Some systems are however both hard to simulate and very sensitive to small perturbations. An additional difficulty arises when a RL agent is…

Holy Grail Math 6.5 Rigor 5.5
Costs: UnclearOut-of-sample: UnclearResolution: L2 Order Book
More details
Authors
Vincent Ragel, Damien Challet
Evaluation
Trading P&L

Browse

All authors · Research topics · Papers with code · Download the scored dataset

📬 The Quant Space Weekly

One email a week: the most interesting quant finance papers, scored and summarized. No spam, unsubscribe anytime.