Statistical arbitrage trading on the intraday market using the asynchronous advantage actor–critic method

In this paper, we focus on statistical arbitrage trading opportunities involving the continuous exploitation of price differences arising during an intraday period with option closing positions balancing market. We aim to maximise reward–risk ratio autonomous strategy. To find optimal policy, propose utilising asynchronous advantage actor–critic (A3C) algorithm, a deep reinforcement learning method, function approximators two-headed shared neural networks. enforce risk-constrained strategy by limiting maximum allowed position, and conduct state engineering selection processes. introduce novel reward goal-based exploration, i.e. behaviour cloning. Our methodology is evaluated case study using limit order book European single coupled market (SIDC) available for Dutch area. The majority hourly products test set return profit. expect our benefit electricity traders, renewable producers researchers who seek implement state-of-art intelligent strategies.