Independent Component Analysis for Three-Way Data With an Application From Atmospheric Science

In this paper, a new approach to independent component analysis (ICA) for three-way data is considered. The rotational freedom of the three-mode component analysis (Tucker3) model is exploited to implement ICA in one mode of the data. The performance of the proposed approach is evaluated by means of numerical experiments. An illustration with real data from atmospheric science is presented, where the first mode is spatial location, the second is time and the third is a set of different meteorological variables representing geopotential heights at various vertical pressure levels. The results show that the three-mode decomposition finds spatial patterns of climate anomalies which can be interpreted in a meteorological sense and as such gives an insightful low-dimensional representation of the data.

[1]  Richard A. Harshman,et al.  Foundations of the PARAFAC procedure: Models and conditions for an "explanatory" multi-model factor analysis , 1970 .

[2]  Nadège Martiny,et al.  Characterization of the Interannual and Intraseasonal Variability of West African Vegetation between 1982 and 2002 by Means of NOAA AVHRR NDVI Data , 2007 .

[3]  Maurizio Vichi.  New Developments in Classification and Data Analysis , 2005 .

[4]  A. Stegeman.  Comparing Independent Component Analysis and the Parafac model for artificial multi-subject fMRI data , 2007 .

[5]  Gene H. Golub,et al.  Matrix computations (3rd ed.) , 1996 .

[6]  Ian T. Jolliffe,et al.  Independent Component Analysis of Climate Data: A New Look at EOF Rotation , 2009 .

[7]  H. Kiers.  Joint Orthomax Rotation of the Core and Component Matrices Resulting from Three-mode Principal Components Analysis , 1998 .

[8]  D. Stephenson,et al.  Observed Trends and Teleconnections of the Siberian High: A Recently Declining Center of Action , 2005 .

[9]  J. Hurrell,et al.  An overview of the North Atlantic Oscillation , 2013 .

[10]  Heng Tao Shen,et al.  Principal Component Analysis , 2009, Encyclopedia of Biometrics.

[11]  Seungjin Choi,et al.  Independent Component Analysis , 2009, Handbook of Natural Computing.

[12]  I. Jolliffe,et al.  Independent exploratory factor analysis with application to atmospheric science data , 2010 .

[13]  Rasmus Bro,et al.  The N-way Toolbox for MATLAB , 2000 .

[14]  Pieter M. Kroonenberg,et al.  An efficient algorithm for TUCKALS3 on data with large numbers of observation units , 1992 .

[15]  Robert H. Halstead,et al.  Matrix Computations , 2011, Encyclopedia of Parallel Computing.

[16]  Sabine Van Huffel,et al.  Imposing Independence Constraints in the CP Model , 2007, ICA.

[17]  S. Mulaik.  Foundations of Factor Analysis , 1975 .

[18]  Bülent Yener,et al.  Unsupervised Multiway Data Analysis: A Literature Survey , 2009, IEEE Transactions on Knowledge and Data Engineering.

[19]  J. Curry,et al.  Encyclopedia of atmospheric sciences , 2002 .

[20]  Richard M. Everson,et al.  ICA: model order selection and dynamic source models , 2001 .

[21]  Peter Filzmoser,et al.  Introduction to Multivariate Statistical Analysis in Chemometrics , 2009 .

[22]  J. Chang,et al.  Analysis of individual differences in multidimensional scaling via an n-way generalization of “Eckart-Young” decomposition , 1970 .

[23]  J. Chisholm.  Approximation by Sequences of Padé Approximants in Regions of Meromorphy , 1966 .

[24]  Ganesh R. Naik,et al.  Introduction: Independent Component Analysis , 2012 .

[25]  J. Leeuw,et al.  Principal component analysis of three-mode data by means of alternating least squares algorithms , 1980 .

[26]  M. Browne.  An Overview of Analytic Rotation in Exploratory Factor Analysis , 2001 .

[27]  Filipe Aires,et al.  Rotation of Eofs by the Independent Component Analysis: Towards a Solution of the Mixing Problem in the Decomposition of Geophysical Time Series , 2013 .

[28]  Rasmus Bro,et al.  A comparison of algorithms for fitting the PARAFAC model , 2006, Comput. Stat. Data Anal..

[29]  F. L. Hitchcock.  The Expression of a Tensor or a Polyadic as a Sum of Products , 1927 .

[30]  H. Kiers.  Three-mode orthomax rotation , 1997 .

[31]  P. Kroonenberg.  Model Selection Procedures in Three-Mode Component Models , 2005 .

[32]  Abdel Hannachi,et al.  Pattern hunting in climate: a new method for finding trends in gridded climate data , 2007 .

[33]  Henk A L Kiers,et al.  A fast method for choosing the numbers of components in Tucker3 analysis. , 2003, The British journal of mathematical and statistical psychology.

[34]  Andrzej Cichocki,et al.  Nonnegative Matrix and Tensor Factorization T , 2007 .

[35]  C. F. Beckmann,et al.  Tensorial extensions of independent component analysis for multisubject FMRI analysis , 2005, NeuroImage.

[36]  Jayanta Basak,et al.  Weather Data Mining Using Independent Component Analysis , 2004, J. Mach. Learn. Res..

[37]  L. Tucker,et al.  Some mathematical notes on three-mode factor analysis , 1966, Psychometrika.

[38]  H. Kiers.  Three-way SIMPLIMAX for oblique rotation of the three-mode factor analysis core to simple structure , 1998 .

[39]  Pieter M. Kroonenberg,et al.  Gram-Schmidt versus Bauer-Rutishauser in alternating least-squares algorithms for three-mode principal component analysis , 1989 .

[40]  H. Nakamura,et al.  A Reexamination of the Northern Hemisphere Sea Level Pressure Variability by the Independent Component Analysis , 2006 .

[41]  J. Kruskal.  Three-way arrays: rank and uniqueness of trilinear decompositions, with application to arithmetic complexity and statistics , 1977 .

[42]  James W. Hurrell,et al.  The North Atlantic Oscillation: Climate Significance and Environmental Impact , 2003 .

[43]  Filipe Aires,et al.  Independent component analysis of multivariate time series: Application to the tropical SST variability , 2000 .

[44]  R. Jennrich,et al.  Independent component analysis as a rotation method: a very different solution to Thurstone's box problem. , 2005, The British journal of mathematical and statistical psychology.

[45]  F. L. Hitchcock.  Multiple Invariants and Generalized Rank of a P‐Way Matrix or Tensor , 1928 .

[46]  P. Kroonenberg.  Applied Multiway Data Analysis , 2008 .

[47]  H. Kiers.  Towards a standardized notation and terminology in multiway analysis , 2000 .

[48]  Pieter M. Kroonenberg,et al.  Three-mode principal component analysis : theory and applications , 1983 .

[49]  D. Ruppert.  The Elements of Statistical Learning: Data Mining, Inference, and Prediction , 2004 .

[50]  Robert I. Jennrich,et al.  A simple general procedure for orthogonal rotation , 2001 .

[51]  H. Kiers.  TUCKALS core rotations and constrained TUCKALS modelling , 1992 .

[52]  Richard M. Everson,et al.  Independent Component Analysis: Principles and Practice , 2001 .

[53]  H. Kiers,et al.  Three-mode principal components analysis: choosing the numbers of components and sensitivity to local optima. , 2000, The British journal of mathematical and statistical psychology.

[54]  Mark D. Plumbley,et al.  Proceedings of the 7th international conference on Independent component analysis and signal separation , 2007 .

[55]  James W. Hurrell,et al.  The North Atlantic Oscillation: Climatic Significance and Environmental Impact , 2003 .

[56]  Joos Vandewalle,et al.  Computation of the Canonical Decomposition by Means of a Simultaneous Generalized Schur Decomposition , 2005, SIAM J. Matrix Anal. Appl..

[57]  Ian T. Jolliffe,et al.  Empirical orthogonal functions and related techniques in atmospheric science: A review , 2007 .