投资时间序列的混沌分维研究


Autoria(s): 刘永熙; 闻邦椿; 杨积东; 郭立新
Data(s)

2001

Resumo

以某市固定资产投资同比增长数据为根据 ,利用混沌动力学中的相空间重构技术、幅值谱分析、庞加莱截面等方法进行分析 ,并计算其关联分维 ,得出其饱和嵌入维数为 5、关联分维约为 1 7,初步确定某市固定资产投资增长过程具有混沌特性 ,有 5个因素影响着某市固定资产投资的同比增长幅度 ,其中 2个因素起决定作用·该法可用于复杂的经济系统的分析研究·

On the basis of the monthly reported data of fixed capital investments in a city, the phase space reconstruction technique of chaotic dynamics, amplitude spectra and Poincar map analysis method were used to study the fixed capital investments. The correlation fractal dimension was calculated. The saturation embed dimension is 5,and the correlation fractal dimension is 1  7. The fixed capital investments process of the city is chaotic. The investment rising is affected by five factors,and two of them are determinant. A new method to investigate the complicated economic system was developed.

国家自然科学基金资助重大项目 ( 19990 5 10 )

Identificador

http://ir.sia.ac.cn//handle/173321/3319

http://www.irgrid.ac.cn/handle/1471x/171850

Idioma(s)

中文

Palavras-Chave #固定资产投资 #时间序列 #非线性机制 #庞加莱截面 #关联分维 #相空间重构 #混沌
Tipo

期刊论文