High-dimensional delay selection for regression models with mutual information and distance-to-diagonal criteria
作者： Geoffroy Simon , Michel Verleysen
DOI: 10.1016/J.NEUCOM.2006.10.150
关键词:
摘要: Delay selection for time series phase space reconstruction may be performed using a mutual information (MI) criterion. However, the delay is in that case limited to estimation of single MI between two variables only. A high-dimensional estimator used select more than one but this approach rather consuming. In paper, an alternative fast criterion proposed optimize all delays reconstruction: distance-to-diagonal (DD) criterion, based on geometrical heuristic. The use distance diagonal illustrated and compared artificial benchmark series.
elsevier.com
sci-hub.se 参考文章(21)
Yongnan Ji, Jin Hao, Nima Reyhani, Amaury Lendasse, Direct and Recursive Prediction of Time Series Using Mutual Information Selection Computational Intelligence and Bioinspired Systems. pp. 1010- 1017 ,(2005) , 10.1007/11494669_124
Holger Kantz, Thomas Schreiber, Nonlinear time series analysis ,(1997)
Floris Takens, Detecting strange attractors in turbulence Lecture Notes in Mathematics. ,vol. 898, pp. 366- 381 ,(1981) , 10.1007/BFB0091924
Christopher M. Bishop, Neural networks for pattern recognition ,(1995)
Antti Sorjamaa, Amaury Lendasse, Jin Hao, Mutual information and k-nearest neighbors approximator for time series prediction international conference on artificial neural networks. pp. 553- 558 ,(2005) , 10.1007/11550907_87
Antti Sorjamaa, Nima Reyhani, Amaury Lendasse, Input and Structure Selection for k-NN Approximator Computational Intelligence and Bioinspired Systems. pp. 985- 992 ,(2005) , 10.1007/11494669_121
Amaury Lendasse, Nima Reyhani, Yongnan Ji, Jin Hao, Mutual Information and Gamma Test for Input Selection the european symposium on artificial neural networks. pp. 503- 508 ,(2005)
M. Orr, Optimising the widths of radial basis functions brazilian symposium on neural networks. pp. 26- 29 ,(1998) , 10.1109/SBRN.1998.730989
Geoffroy Simon, Michel Verleysen, Lag selection for regression models using high-dimensional mutual information the european symposium on artificial neural networks. pp. 395- 400 ,(2006)
Nabil Benoudjit, Michel Verleysen, On the Kernel Widths in Radial-Basis Function Networks Neural Processing Letters. ,vol. 18, pp. 139- 154 ,(2003) , 10.1023/A:1026289910256