Advertisement

An Algorithm for Finding Projections with Extreme Kurtosis

  • Cinzia Franceschini
  • Nicola Loperfido
Conference paper
Part of the Springer Proceedings in Mathematics & Statistics book series (PROMS, volume 227)

Abstract

Projection pursuit is a multivariate statistical technique aimed at finding interesting low-dimensional data projections. A projection pursuit index is a function which associates a data projection to a real value measuring its interestingness: the higher the index, the more interesting the projection. Consequently, projection pursuit looks for the data projection which maximizes the projection pursuit index. The absolute value of the fourth standardized cumulant is a prominent projection pursuit index. In the general case, a projection achieving either minimal or maximal kurtosis poses computational difficulties. We address them by an algorithm which converges to the global optimum, whose computational advantages are illustrated with air pollution data.

Keywords

Fourth moment Kurtosis Projection pursuit Tensor 

Notes

Acknowledgements

The authors would like to thank an anonymous referee for her/his comments which greatly helped in improving the present paper.

References

  1. 1.
    Adcock, C.J., Shutes, K.: On the multivariate extended skew-normal, normal-exponential, and normal-gamma distributions. J. Stat. Theory Pract. 6, 636–664 (2012)MathSciNetCrossRefGoogle Scholar
  2. 2.
    Bartoletti, S., Loperfido, N.: Modelling air pollution data by the skew-normal distribution. Stoch. Environ. Res. Risk Assess. 24, 513–517 (2010)CrossRefGoogle Scholar
  3. 3.
    Caussinus, H., Ruiz-Gazen, A.: Exploratory Projection Pursuit. In: Gerard Govaert, G. (ed.) Data Analysis, pp. 76–92. Wiley (2009)Google Scholar
  4. 4.
    de Lathauwer, L., Comon, P., De Moor, B., Vandewalle, J.: Higher-order power method Application in independent component analysis. In: Proceedings of the International Symposium on Nonlinear Theory and Its Applications (NOLTA’95), Las Vegas, NV, pp. 91–96 (1995)Google Scholar
  5. 5.
    de Lathauwer L., de Moor, B., Vandewalle, J.: On the best rank-1 and rank-(\(R_{1}\),\(R_{2}\),...\(R_{N}\)) approximation of high-order tensors. SIAM J. Matrix Ana. Appl. 21, 1324–1342 (2000)Google Scholar
  6. 6.
    Hou, S., Wentzell, P.D.: Fast and simple methods for the optimization of kurtosis used as a projection pursuit index. Anal. Chim. Acta 704, 1–15 (2011)CrossRefGoogle Scholar
  7. 7.
    Huber, P.J.: Projection pursuit (with discussion). Ann. Stat. 13, 435–525 (1985)CrossRefzbMATHGoogle Scholar
  8. 8.
    Hyvarinen, A., Oja, E.: A fast fixed point-algorithm for independent component analysis. Neural Comput. 9, 1483–1492 (1997)CrossRefGoogle Scholar
  9. 9.
    Hyvarinen, A., Karhunen, J., Oja, E.: Independent Compon. Anal. Wiley, New York (2001)CrossRefGoogle Scholar
  10. 10.
    Jondeau, E., Rockinger, M.: Optimal portfolio allocation under higher moments. Eur. Finan. Manag. 12, 29–55 (2006)CrossRefGoogle Scholar
  11. 11.
    Kent, J.T.: Discussion on the paper by Tyler, Critchley, Dumbgen & Oja: invariant co-ordinate selection. J. R. Stat. Soc. Ser. B 71, 575–576 (2009)Google Scholar
  12. 12.
    Kofidis, E., Regalia, P.A.: On the best rank-1 approximation of higher-order supersymmetric tensors. SIAM J. Matrix Anal. Appl. 23, 863–884 (2002)MathSciNetCrossRefzbMATHGoogle Scholar
  13. 13.
    Loperfido, N.: Spectral analysis of the fourth moment matrix. Linear Algebra Appl. 435, 1837–1844 (2011)MathSciNetCrossRefzbMATHGoogle Scholar
  14. 14.
    Loperfido, N.: A new Kurtosis matrix, with statistical applications. Linear Algebra Appl. 512, 1–17 (2017)MathSciNetCrossRefzbMATHGoogle Scholar
  15. 15.
    Paajarvi, P., Leblanc, J.P.: Skewness maximization for impulsive sources in blind deconvolution. In: Proceedings of the 6th Nordic Signal Processing Symposium—NORSIG 2004, 9–11 June, Espoo, Finland (2004)Google Scholar
  16. 16.
    Peña, D., Prieto, F.J.: Multivariate outlier detection and robust covariance estimation (with discussion). Technometrics 43, 286–310 (2001)Google Scholar
  17. 17.
    Peña, D., Prieto, F.J.: Cluster identification using projections. J. Am. Stat. Assoc. 96, 1433–1445 (2001)Google Scholar
  18. 18.
    Peña, D., Prieto, F.J., Viladomat, J.: Eigenvectors of a kurtosis matrix as interesting directions to reveal cluster structure. J. Multiv. Anal. 101, 1995–2007 (2010)MathSciNetCrossRefzbMATHGoogle Scholar
  19. 19.
    Qi, L.: Rank and eigenvalues of a supersymmetric tensor, the multivariate homogeneous polynomial and the algebraic hypersurface it defines. J. Symb. Comput. 41, 1309–1327 (2006)MathSciNetCrossRefzbMATHGoogle Scholar
  20. 20.
    Ruiz-Gazen A., Marie-Sainte S.L., Berro, A.: Detecting multivariate outliers using projection pursuit with Particle Swarm Optimization. In: Proceedings of COMPSTAT, pp. 89–98 (2010)Google Scholar
  21. 21.
    Zarzoso, V., Comon, P.: Robust independent component analysis by iterative maximization of the Kurtosis contrast with algebraic optimal step size. IEEE Trans. Neural Netw. 21, 248–261 (2010)CrossRefGoogle Scholar

Copyright information

© Springer International Publishing AG, part of Springer Nature 2018

Authors and Affiliations

  1. 1.Dipartimento di Economia (DEC)Università degli Studi “G. d’Annunzio” di Chieti-PescaraPescaraItaly
  2. 2.Dipartimento di Economia, Società e Politica (DESP)Università degli Studi di Urbino “Carlo Bo”UrbinoItaly

Personalised recommendations