Iron Ore Futures Trading Strategy Based on EEMD and DRL
With the financialization of iron ore and other commodities,more and more investors are involved in iron ore fu-tures trading.The trading strategy has become a key research topic in investment decision-making.Due to the volatile price fluctuations of iron ore futures,an iron ore futures trading strategy is designed based on ensemble empirical mode decomposition(EEMD)and deep reinforcement learning(DRL).Firstly,the EEMD method is used to decompose the characteristics of iron ore futures price,and the trading environment of iron ore futures based on Markov decision process is constructed by considering the characteristics after decomposition.Secondly,a variety of DRLs are used to generate iron ore futures trading strategies,which are optimized by the cumulative return.Finally,the Sharpe ratio is used to screen out the optimal strategy in each trading cycle,and the optimal strategy combination in the whole trading cycle is formed.The experimental results show that the proposed strate-gy has strong robustness based on ensuring the maximization of return.
万方数据
维普
