A Decision Support System for Trading in Apple Futures Market Using Predictions Fusion

In the last decade, High-Frequency Trading (HFT) has become a popular issue in futures market, which attracted much attention from numerous researchers. this study, an intelligent decision support system is proposed for apple high-frequency trading. First, three eXtreme Gradient Boosting (XGBoost) based models use feature inputs multiple time scales return and direction prediction. Then, on pre-designed trading rule, signals of long short-selling are determined, corresponding transactions executed. order to retain considerable profits avoid serious losses possibly caused by sudden huge price changes toward opposite as predictions, position closing function implemented rule. Meanwhile, Particle Swarm Optimization (PSO) employed optimize parameters rule well XGBoost parameters. By evaluating experimental results, we observed that approach successfully achieved best performance terms prediction accuracy, transaction returns, return/risk ratio. It could be inferred results provide beneficial reference market traders involved futures.