MEASURING NONLINEAR DEPENDENCE BETWEEN TIME SERIES
-
摘要: 基于分数维数,提出了非线性相关度的概念,用于度量两列经济数据的非线性相关程度,以解决非线性经济预测中的变量选择问题.仿真结果与应用结果说明该方法效果较好.Abstract: In this paper, a statistic-Nonlinear Dependence Coefficient based on fractal dimension is proposed. This statistic is designed for measuring nonlinear dependence between time series, and can be used in variable selecting for economical forecasting. Numerical results show the statistic is reasonable.
-
Keywords:
- nonlinear dynamics /
- fractal dimension /
- forecasting /
- statistic
-
-
[1] Takens F. Detecting Strange Attractors in Turbulence, in:Dynamical Systems and Turbulence, Warwick 1980, Lecture Notes in Mathematics 898, Springer-Verlag, Berlin [2] Grassberger P, Procaccia I. Measuring the Strangeness of Strange Attractors. Phys. 9D, 1983 [3] Brock W A. Distinguishing Random and Deterministic Systems:Abridged Version. Journal of Economic Theory 40, 1986 [4] Brock W, Dechert W D, Scheinkman J. A Tests for Independence Based on the Correlation Dimension. Working Paper,Department of Economics, University of Wisconsin, Madison, 1987 [5] Henon M. A Two-dimensional Mapping with a Strange Attractor. Communacations in Mathematical Physics 1976,50:69~77 [6] 张朋拄,曹文华,樊重俊,韩崇昭.金川有色金属公司经济活动分析经营计划决策支持系统的快速生成-UCP-IDSS生成环境应用之二.决策与决策支持系统,1995,5(3) [7] Fan Chongjun, Han Chongzhao, Hu Baosheng. Forecasting the Behavior of Multivariate Time Series with Neural Networks in a DSS for Business Planning. Proceedings of ISIM'96 (Nanjing,1996)
计量
- 文章访问数: 830
- HTML全文浏览量: 0
- PDF下载量: 162