基于移动平均线的趋势跟踪交易策略研究

发布时间：2018-01-11 01:30

本文关键词：基于移动平均线的趋势跟踪交易策略研究　出处：《上海交通大学》2014年硕士论文　论文类型：学位论文

【摘要】：算法交易(广义)经过了三十多年的发展,现在欧美国家金融市场已得到广泛应用。算法交易分为决策算法和执行算法。决策算法主要是解决何时买(卖)以及买(卖)什么的问题,而执行算法主要是解决如何买(卖)的问题。本文主要研究对象是决策算法中的一种典型的策略——趋势跟踪策略。趋势跟踪这一技术交易策略在国内外都得到了充分的研究和应用,并被证明其有效性。本文首先对算法交易的背景及概念做了系统介绍后,详细分析了典型的决策算法交易模型,为后文的模型设计做了铺垫。然后,利用自相关系数的分析,发现沪深300指数在中短期的持续性和长期的均值回归特征。本文以沪深300指数十年的交易数据作为标的进行阶段的以及全程的回测,对以技术分析为基础的传统的趋势跟踪策略进行分析,对比了简单移动平均线和指数移动平均线在这一策略中应用的效果,发现经过指数平滑的移动平均线比简单移动平均线更有效。之后在传统策略的基础上进行改进,加入了止损规则和仓位控制规则。仓位控制是通过经处理的均线差值所蕴含的趋势强度进行的,根据趋势的强弱控制投资的多少。止损规则能有效的减少亏损并提高收益,仓位控制规则有很好的降低风险和减少亏损的效果,尤其对于日波动率大的样本区间。该策略在样本以及样本外测试中都获得了高于市场的较好收益,在跨样本区间的测试验证了稳定性。本文对于趋势跟踪策略方法的研究以及投资者应用都有较高价值。
[Abstract]:Algorithmic transaction (generalized) has been developed for more than 30 years. Now financial markets in Europe and the United States have been widely used. Algorithm transactions are divided into decision algorithm and execution algorithm. Decision algorithm is mainly to solve the problem of when to buy (sell) and what to buy (sell). And the execution algorithm is mainly to solve how to buy (sell). The main research object of this paper is a typical strategy of decision algorithm, trend tracking, which has been fully studied and applied both at home and abroad. This paper firstly introduces the background and concept of the algorithm transaction, then analyzes the typical decision algorithm transaction model in detail, which lays the groundwork for the later model design. Using the analysis of autocorrelation coefficient. We find the persistence of the CSI 300 index in the short and medium term and the long-term mean regression characteristics. This paper takes the trading data of CSI 300 years as the target to carry on the stage and the whole process of backtracking. The traditional trend tracking strategy based on technical analysis is analyzed, and the effect of simple moving average and exponential moving average in this strategy is compared. It is found that the moving average after exponential smoothing is more effective than the simple moving average, and then it is improved on the basis of the traditional strategy. The stop loss rule and position control rule are added. Position control is carried out through the trend strength contained in the processed mean line difference. According to the trend of the strength of the control investment. Stop loss rules can effectively reduce losses and improve earnings, position control rules have a good effect of reducing risk and loss. Especially for the sample interval with high daily volatility, the strategy has better returns than the market in both sample and out-of-sample tests. The stability is verified in the cross-sample interval test. This paper has a high value for the research of trend tracking strategy and the application of investors.
【学位授予单位】：上海交通大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F832.51

【相似文献】