基于多目标优化的股市趋势追踪交易策略

Trend Following Trading Strategy Based on Multi-Objective Optimization in Stock Market

针对股票市场交易中交易时机的不确定性及不稳定性,提出了一种基于多目标优化的股市趋势追踪交易策略。在提出的交易策略中,一共使用三对基于移动平均收敛发散柱状图线的阈值对股票的买卖时机进行预测。这三对阈值的作用分别是:第一对阈值判断股票微弱的上涨趋势和下跌趋势;第二对阈值检测股票暴涨和暴跌的情况;第三对阈值在移动平均收敛发散柱状图线的中心线判断买入点和卖出点。本文改进用于预测的股市模型,改进了判定股票暴增或暴跌的方法,并改进了阈值的搜索空间范围,比较了不同多目标优化算法对阈值的优化效果。最后与前人的静态算法、自适应算法、基于移动平均收敛发散线的多阈值算法效果进行对比,本文提出的方法获利成功率分别提升了10%、7%和12%,累计投资回报率分别提升了7.09%、10.49%和3.98%。

In order to alleviate the uncertainty and instability of trading timing in stock market transactions,a stock trend following trading strategy based on multi-objective optimization is proposed.In the proposed trading strategy,a total of three pairs of thresholds based on the moving average convergence divergence histogram line(MACDHL)are used to predict the timings of stock trading.The first pair of thresholds is used to judge the weak upward trend and downward trend of stocks,the second pair of thresholds is used to detect the situation of stocks’boom and crash,and the third pair of thresholds judge the buying points and selling points at the center line of MACDHL.This paper perfects the stock market model used for forecasting,improves the method of judging the stock boom or crash,and improves the search space range of the thresholds,and compares the optimization effect of different multi-objective optimization algorithms on these thresholds.Finally,compared with the previous static algorithms,adaptive algorithms and multi-threshold algorithms based on the Moving average convergence divergence line,the profitable success rate of the method proposed in this paper has increased by 10%,7%and 12%,and accumulated return on investment has increased by 7.09%,10.49%and 3.98%.