基于EMD与STSA混合方法的金融收益信息提取与预测  

Extraction and Forecasting of Financial Returns Based on Hybrid Method of EMD and STSA

作　　者：李祥飞[1] 张再生[1] 黄超[2] 高杨[1] 

机构地区：[1]天津大学管理与经济学部,天津300072 [2]上海财经大学会计学院,上海200433

出　　处：《系统工程》2014年第2期138-144,共7页Systems Engineering

基　　金：国家自然科学基金资助项目(70971097)

摘　　要：针对STSA方法在金融时间序列分析中的缺陷提出了运用EMD与STSA结合的改进方法。以上证指数、深证成指、建筑指数、金融指数、地产指数、上证商业6种指数的收益数据作为研究样本,利用EMD方法分解提取出一系列反映原始序列不同时间尺度信息的分量,通过对各分量进行STSA分析后发现导致原始序列多变化模式的原因。在此基础上提出了通过单一变化模式分量对原始序列变化趋势进行估计的条件和限定范围,实验结果表明,该方法在提取和分析时间序列变化模式方面具有独特的优势,具有较高的预测精度和实用性。This paper puts forward an improved method by combined EMD and STSA together direct at the defect of STSA method in financial time series analysis. Taking the return data of 6 different stock indexes as the research sample and using EMD decomposition method extracted a series of signal components which reflect different time scale information of the original sequence. By using STSA to each component, we have found the reason leading to the diverse changing pattern of the original sequence. Based on this, we propose a method that using the single changing pattern component to estimate the original sequence, the condition and limited range are given as well. The experimental results show that the proposed approach has uniquely superiority in extract the information of time series, and is of high precision and practicability as well.

关 键 词：经验模态分解 符号时间序列 信息提取 预测 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]