Machine learning for trading

Gordon Ritter applies reinforcement learning to dynamic trading strategies with market impact

CLICK HERE TO VIEW THE PDF

In multiperiod trading with realistic market impact, determining the dynamic trading strategy that optimises the expected utility of final wealth can be difficult. Gordon Ritter shows that, with an appropriate choice of reward function, reinforcement learning techniques (specifically Q-learning) can successfully handle the risk-averse case

In this article, we show how machine learning can be applied to the problem of discovering and implementing dynamic trading