Azad, Rajeev K. ; Subba Rao, J. ; Ramaswamy, Ramakrishna (2004) Symbol sequence analysis of climatic time signals Nonlinear Analysis: Real World Applications, 5 (3). pp. 487-500. ISSN 1468-1218
Full text not available from this repository.
Official URL: http://www.sciencedirect.com/science/article/pii/S...
Related URL: http://dx.doi.org/10.1016/j.nonrwa.2003.11.003
Abstract
The conditional entropy of two symbolic sequences encoded faithfully from different chaotic time signal data is minimised at zero relative shift of the two signals if the signals have their origin in same underlying dynamics. We show that this minimum in conditional entropies E(Z/X) and E(X/Z) obtained after suitably 'coarse-graining' the time signals of, say, variables X and Z have marked differences depending upon the degree of dynamical coupling of the variables in the model. This technique has also been shown to be useful in studying delayed dependences of time signals. Application is made to climate data of four meteorological stations in India, namely New Delhi, Jaipur, Sundernagar and Chennai in order to determine (a) the commonality of underlying dynamics, (b) relative strength of dynamical coupling of different variables and (c) the delay implicit in the dynamics. The method appears robust to measurement noise.
Item Type: | Article |
---|---|
Source: | Copyright of this article belongs to Elsevier Science. |
Keywords: | Coarse-graining; Information Entropy; Dynamical Coupling; Climate Data; Time-delays |
ID Code: | 45356 |
Deposited On: | 28 Jun 2011 04:51 |
Last Modified: | 28 Jun 2011 04:51 |
Repository Staff Only: item control page