An algorithm for arbitrary–order cumulant tensor calculation in a sliding window of data streams

Abstract High-order cumulant tensors carry information about statistics of non-normally distributed multivariate data. In this work we present a new efficient algorithm for calculation of cumulants of arbitrary orders in a sliding window for data streams. We show that this algorithm offers substantial speedups of cumulant updates compared with the current solutions. The proposed algorithm can be used for processing on-line high-frequency multivariate data and can find applications, e.g., in on-line signal filtering and classification of data streams. To present an application of this algorithm, we propose an estimator of non-Gaussianity of a data stream based on the norms of high order cumulant tensors. We show how to detect the transition from Gaussian distributed data to non-Gaussian ones in a data stream. In order to achieve high implementation efficiency of operations on super-symmetric tensors, such as cumulant tensors, we employ a block structure to store and calculate only one hyper-pyramid part of such tensors.

[1]  Laurent Albera,et al.  Localization of extended brain sources from EEG/MEG: The ExSo-MUSIC approach , 2011, NeuroImage.

[2]  Alan Edelman,et al.  Julia: A Fast Dynamic Language for Technical Computing , 2012, ArXiv.

[3]  D. Cox,et al.  Asymptotic techniques for use in statistics , 1989 .

[4]  Bertrand Maillet,et al.  Multi-moment Asset Allocation and Pricing Models , 2006 .

[5]  Eric Jondeau,et al.  Moment Component Analysis: An Illustration With International Stock Markets , 2015 .

[6]  Jesús S. Aguilar-Ruiz,et al.  Knowledge discovery from data streams , 2009, Intell. Data Anal..

[7]  Ronald L. Graham,et al.  Concrete mathematics - a foundation for computer science , 1991 .

[8]  Ian W. R. Martin,et al.  Consumption-Based Asset Pricing with Higher Cumulants , 2010 .

[9]  Monte Carlo Approximate Tensor Moment Simulations , 2014 .

[10]  K. Nordhausen,et al.  Joint Use of Third and Fourth Cumulants in Independent Component Analysis , 2015, 1505.02613.

[11]  Krzysztof Krawiec,et al.  Exploring complex and big data , 2017, Int. J. Appl. Math. Comput. Sci..

[12]  Alan Edelman,et al.  Array Operators Using Multiple Dispatch: A design methodology for array implementations in dynamic languages , 2014, ARRAY@PLDI.

[13]  E. Luciano,et al.  Copula methods in finance , 2004 .

[14]  Tze Meng Low,et al.  Exploiting Symmetry in Tensors for High Performance: Multiplication with Symmetric Tensors , 2013, SIAM J. Sci. Comput..

[15]  A. Ferreol,et al.  High-Resolution Direction Finding From Higher Order Statistics: The$2rm q$-MUSIC Algorithm , 2006, IEEE Transactions on Signal Processing.

[16]  Laurenz Wiskott,et al.  CuBICA: independent component analysis by simultaneous third- and fourth-order cumulant diagonalization , 2004, IEEE Transactions on Signal Processing.

[18]  Liqun Qi,et al.  Eigenvalues of a real supersymmetric tensor , 2005, J. Symb. Comput..

[19]  Ronald L. Graham,et al.  Concrete Mathematics, a Foundation for Computer Science , 1991, The Mathematical Gazette.

[20]  F Wendling,et al.  EEG extended source localization: Tensor-based vs. conventional methods , 2014, NeuroImage.

[21]  Aapo Hyvärinen Independent Component Analysis of Images , 2014, Encyclopedia of Computational Neuroscience.

[22]  Cumulant filters - a recursive estimation method for systems with non-Gaussian process and measurement noise , 2003, Proceedings of the 35th Southeastern Symposium on System Theory, 2003..

[23]  Maurice G. Kendall,et al.  The advanced theory of statistics , 1945 .

[24]  Piotr Gawron,et al.  Efficient Computation of Higher-Order Cumulant Tensors , 2017, SIAM J. Sci. Comput..

[25]  Alan Edelman,et al.  Julia: A Fresh Approach to Numerical Computing , 2014, SIAM Rev..