About this document
Reinforcement Learning in Statistical Arbitrage by chris is a document available to read on EtoBox.
The document discusses using reinforcement learning to develop a model-free framework for statistical arbitrage trading. It introduces using an empirical mean reversion time metric to construct mean-reverting spreads without relying on assumptions. A reinforcement learning algorithm is also used to dynamically optimize trading decisions over time instead of using predefined rules and parameters estimated during portfolio formation.
- Author
- chris
- Language
- EN