Deep Reinforcement Learning Approach to Portfolio Optimization in the Australian Stock Market

The future of portfolio management is evolving from relying on human expertise to incorporating artificial intelligence techniques. Traditional techniques such as fundamental and technical analysis will eventually be replaced by more sophisticated Deep Reinforcement Learning (DRL) algorithms. However, it is still a long way from designing a profitable strategy in the complex and dynamic stock market. While previous studies have focused on the American stock market, this paper applies two DRL algorithms: Proximal Policy Optimization (PPO) and Advantage Actor-Critic (A2C) to trade the constituent stocks of Australian Securities Exchange 50 Index (ASX50).This study also introduces a weighted moving average to the action space and a transaction threshold to help the agents reduce trivia trades that result in high transaction costs. The results are presented and benchmarked against the ASX50 Index.In order to reduce the frequency and volume of trading made by the agents, we introduced a weighted moving average and a transaction threshold to determine if the trade executed by the agents is necessary. The weighted moving average and threshold was able to help reduce the number of trivia trades made by the agent. The optimal window for the moving average and the optimal threshold were determined to be 15 and 0.04 for PPO and 15 and 0.05 for A2C respectively.The A2C agent was better at following trends and had the higher upside potential but can suffer from more severe damage during bearish markets. On the other hand, the PPO agent had the lowest annual volatility and highest maximum drawdown, which is more helpful in a bearish or volatile market