K线能量计算的股市生命期态势预测方法  

K line energy calculation method for stock market lifetime prediction

作　　者：姚宏亮[1] 周光辉[1] 李俊照[1] 

机构地区：[1]合肥工业大学计算机与信息学院,合肥230009

出　　处：《计算机应用研究》2016年第6期1637-1641,1647,共6页Application Research of Computers

基　　金：国家自然科学基金资助项目(61175051;61070131;61175033)

摘　　要：股市中K线特征是股价涨跌的因果信息,基于支持向量机(SVM)的股价预测模型没有考虑K线特征知识,对于股价态势难以有效预测。提出基于K线能量计算的股市生命期支持向量机态势预测算法(LPFSVM),首先,提取典型K线特征,通过引入特征的孕育成熟度和爆发力定义,给出K线特征支持向量机算法(KLF-SVM);进而,在KLF-SVM算法基础上定义特征的能量计算模型给出一种K线能量计算的SVM股价预测算法。为了有效地预测态势,引入股价波动的生命期概念,通过K线组合特征判定股价所处的生命期的阶段,进而结合生命期阶段之间的时序影响关系给出一种基于生命期的股价态势预测算法。在上证和深证数据集上的实验结果表明,LPF-SVM算法对于股价上升波段和下跌波段的股价预测取得了很好的效果。In the stock market,K line feature is the causal information for the rise and fall of stock price. Stock price prediction model of support vector machine（ SVM）,which does not consider K line features,can not predict the stock trend effectively. Based on K line energy calculation,this paper put forward a lifetime support vector machine（ LPF-SVM） algorithm,forecasting stock market situation. First,it extracted the typical K line,with the introduction of maturity and explosive force definition,obtained the K line feature of support vector machine algorithm（ KLF-SVM）. Then on the basis of KLF-SVM,it defined typical energy calculation model,gave a kind of SVM prediction algorithm for K line energy calculation. In order to predict the situation effectively,it introduced the lifetime concept of stock price volatility. The stage of lifetime of the stock price could be judged through the K line combination features,and then combing with the timing effect relationship of life stage,it gave a stock price trend prediction algorithm based on lifetime. The experimental results on the Shanghai and Shenzhen data set show that,the LPF-SVM algorithm can predict the rising and falling band of stock price effectively.

关 键 词：K线特征 孕育成熟度 爆发力 能量 股市生命期 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]