摘要
以上海股票市场为例,运用多重分形谱方法对其多重分形性进行了实证研究。分析了多重分形谱的两个参数(分形谱的宽度及最大、最小概率子集分形维数的差)与指数变化趋势的关系。并进一步研究了多重分形谱参数与时间序列对数收益率的关系。结果表明多重分形谱参数与指数的变化趋势及对数收益率具有明显的关联性,这说明多重分形分析能够揭示更多关于市场变化的有用信息,有望以一定概率预测股价的涨落。
Taking the Shanghai stock market as example, the multifractal property of stock price index was studied. The correlation of two parameters of the multifractal spectra with the variation of closing index and the correlation of two parameters with logarithmic return were studied statistically. It is found that two parameters are correlated with the variation of closing index, and the correlation of two parameters and logarithmic return are obvious. This suggests that mulifractal analysis can reveal more useful information about the market trend and that multifractal analysis can be used to predict the increase or decrease of the stock market index with a larger probability.
出处
《管理评论》
2007年第12期3-8,共6页
Management Review
基金
国家自然科学基金项目资助(70371062)。
