一种计算时间序列关联维的逐步递归法
A Novel Method of Calculating the Correlative Dimension of Time-Series
-
摘要: 该文提出了一种计算时间序列关联维数的逐步递归法。该方法可按给定的置信度和相关性指标,逐步递归、识别并计算出一维时间序列的无标度区及关联维数。理论分析和仿真结果验证了该方法的有效性和客观性。
-
关键词:
- 时间序列; 关联维; 重构相空间; 分形
Abstract: In this paper, a novel method of calculating the correlative dimension of time-series is proposed. With this method, the fractal scaleless band of time-series can be identified and the correlative dimension of time series can be calculated according to the parameters given. The results of theory analysis and Monte Carlo simulation show that the method is efficient and objective in identifying the fractal scaleless band of time-series and calculating the correlative dimension of time series. -
Mandelbrot B B. Fractals: Form, Chance and Dimension. San Francisco: Freeman, 1977.[2]谢和平,薛秀谦.分形应用中的数学基础与方法.北京:科学出版社,1997.[3]Wornell G W, Oppenheim A V. Estimation of fractal of signal from noisy measurement using wavelets[J].IEEE Trans. on SP.1992, 40(3):611-[4]Turcotte D L. Fractals in geology and geophysics[J].Pure and Applied Geophysics.1989, 131:171-[5]Grassberger P, Privacies I. Characterization of strange attractors[J].Physical Review Letters.1983, 50(5):346-[6]Yokoya N, Yamamoto K, Funakubo N. Fractal-based analysis and interpolation of 3D natural surface shapes and their applications to terrain modeling[J].Computer Vision, Graphics and Image Processing.1989, 46:284-[7]汪富泉,罗朝盛,陈国先.关联维数在油气勘探中的应用.中国海上油气(地质),1992,6(6):61-67.[8]Petros Maragos, Fang-Kuo Sun. Measuring the fractal dimension of signals: Morphological covers and iterative optimization[J].IEEE Trans. on SP.1993, 41(1):108- -
计量
- 文章访问数: 2173
- HTML全文浏览量: 77
- PDF下载量: 808
- 被引次数: 0
下载:
