摘要
高频金融数据通常具有尖峰厚尾的非正态特征,常用稳定分布来拟合。本文采用稳定分布的性质对高频上证指数收益率的分布作了标度分析,求得特征指数为1.48,说明我们在分析证券指数分布时可以用稳定分布,而非正态分布。
High frequency finance data are not normal usually, so that stable distribution is usually used to fit them. We apply the character in the stable distribution to the Shanghai Stocks Exchange Composite index, using the scaling of fluctuations, and gain characteristic index, which is 1.48. This shows that we can use stable distribution instead of non-normal one in analyzing securities index.
出处
《运筹与管理》
CSCD
2006年第4期127-129,共3页
Operations Research and Management Science
关键词
高频数据
稳定分布
标度分析
特征指数
high frequency finance data
stable distribution
scaling
characteristic index
