Mining Matrix Data with Bregman Matrix Divergences for Portfolio Selection

  • Richard Nock
  • Brice Magdalou
  • Eric Briys
  • Frank Nielsen


In the early fifties, Markowitz contributed a theory that considerably simplified portfolio choices. He narrowed down the traditional expected utility model and assumed that investors only care for mean and variance. The mean-variance portfolio theory was born. As it name suggests, mean-variance theory is predicated on simple assumptions that are unfortunately seldomly met in real life. Indeed, it is now a well-established fact that for a host of reasons financial returns do not obey Gaussian distributions. This paper first draws on ideas from econometrics, finance and statistics to derive a rigorous generalization of Markowitz’ mean-variance model to a mean-divergence model, lifted to matrix entries, grounded on exponential families of distributions, that we argue is both more realistic and better suited to further developments in learning. The generalized model turns out to heavily rely on Bregman divergences. There has recently been a burst of attention in on-line learning to learn portfolios having limited risk in Markowitz’ setting. In an on-line framework, we then tackle the problem of finding adaptive portfolio strategies based on our generalized model. We devise a learning algorithm based on new matrix generalizations of p-norms to track non stationary target portfolios with limited risk. Theoretical bounds and preliminary experiments over nearly twelve years of S\(\&\)P 500 confirm the validity of the generalized model, the capacity it brings to spot important events that would otherwise be dampened in the mean-variance model, and the potential of the algorithm. Finally, we make an in depth analysis of the matrix divergences and risk premia derived in our model that shed some theoretical light on the ways the risk premium may be blown up as the investor’s portfolio shifts away from a so-called natural market allocation which defines the best (unknown) allocation at the market’s scale.


Risk Aversion Risk Premium Portfolio Selection Matrix Divergence Certainty Equivalent 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.



The authors wish to thank the reviewers for useful comments, and gratefully acknowledge the support of grant ANR-07-BLAN-0328-01.


  1. 1.
    Amari, S.I.: Natural gradient works efficiently in learning. Neural Comput. 10, 251–276 (1998)CrossRefGoogle Scholar
  2. 2.
    Amari, S.I., Nagaoka, H.: Methods of Information Geometry. Oxford University Press, Oxford (2000)Google Scholar
  3. 3.
    Banerjee, A., Guo, X., Wang, H.: On the optimality of conditional expectation as a bregman predictor. IEEE Trans. Inf. Theory 51, 2664–2669 (2005)MathSciNetCrossRefGoogle Scholar
  4. 4.
    Banerjee, A., Merugu, S., Dhillon, I., Ghosh, J.: Clustering with Bregman divergences. J. Mach. Learn. Res. 6, 1705–1749 (2005)MathSciNetzbMATHGoogle Scholar
  5. 5.
    Borodin, A., El-Yaniv, R., Gogan, V.: Can we learn to beat the best stock. In: NIPS*16, pp. 345–352. (2003)Google Scholar
  6. 6.
    Bourguinat, H., Briys, E.: L’Arrogance de la Finance: comment la Théorie Financière a produit le Krach (The Arrogance of Finance: how Financial Theory made the Crisis Worse). La Découverte (2009)Google Scholar
  7. 7.
    Bregman, L.M.: The relaxation method of finding the common point of convex sets and its application to the solution of problems in convex programming. USSR Comp. Math. Math. Phys. 7, 200–217 (1967)CrossRefGoogle Scholar
  8. 8.
    Briys, E., Eeckhoudt, L.: Relative risk aversion in comparative statics: comment. Am. Econ. Rev. 75, 281–283 (1985)Google Scholar
  9. 9.
    Chavas, J.P.: Risk Analysis in Theory and Practice. (Academic Press Advanced Finance) Academic press, London (2004)Google Scholar
  10. 10.
    Cover, T.M.: Universal portfolios. Math. Finance 1, 1–29 (1991)MathSciNetzbMATHCrossRefGoogle Scholar
  11. 11.
    Dhillon, I., Sra, S.: Generalized non-negative matrix approximations with Bregman divergences. In: NIPS*18 (2005)Google Scholar
  12. 12.
    Dhillon, I., Tropp, J.A.: Matrix nearness problems with Bregman divergences. SIAM J. Matrix Anal. Appl. 29, 1120–1146 (2007)MathSciNetCrossRefGoogle Scholar
  13. 13.
    Duchi, J.C., Shalev-Shwartz, S., Singer, Y., Tewari, A.: Composite objective mirror descent. In: Proceedings of the 23\(^{rd}\) COLT, pp. 14–26. (2010)Google Scholar
  14. 14.
    Even-Dar, E., Kearns, M., Wortman, J.: Risk-sensitive online learning. In: 17\(^{th}\) ALT, pp. 199–213. (2006)Google Scholar
  15. 15.
    Kivinen, J., Warmuth, M., Hassibi, B.: The \(p\)-norm generalization of the LMS algorithm for adaptive filtering. IEEE Trans. SP 54, 1782–1793 (2006)CrossRefGoogle Scholar
  16. 16.
    Kulis, B., Sustik, M.A., Dhillon, I.S.: Low-rank kernel learning with Bregman matrix divergences. J. Mach. Learn. Res. 10, 341–376 (2009)MathSciNetzbMATHGoogle Scholar
  17. 17.
    Markowitz, H.: Portfolio selection. J. Finance 6, 77–91 (1952)Google Scholar
  18. 18.
    von Neumann, J., Morgenstern, O.: Theory of games and economic behavior. Princeton University Press, Princeton (1944)Google Scholar
  19. 19.
    Nock, R., Luosto, P., Kivinen, J.: Mixed Bregman clustering with approximation guarantees. In: 23\(^{rd}\) ECML, pp. 154–169. Springer, Berlin (2008)Google Scholar
  20. 20.
    Nock, R., Magdalou, B., Briys, E., Nielsen, F.: On Tracking Portfolios with Certainty Equivalents on a Generalization of Markowitz Model: the Fool, the Wise and the Adaptive. In: Proceedings of the 28\(^{th}\) International Conference on Machine Learning, pp. 73–80. Omnipress, Madison (2011)Google Scholar
  21. 21.
    Ohya, M., Petz, D.: Quantum Entropy and Its Use. Springer, Heidelberg (1993)Google Scholar
  22. 22.
    Petz, D.: Bregman divergence as relative operator entropy. Acta Math. Hungarica 116, 127–131 (2007)MathSciNetzbMATHCrossRefGoogle Scholar
  23. 23.
    Pratt, J.: Risk aversion in the small and in the large. Econometrica 32, 122–136 (1964)zbMATHCrossRefGoogle Scholar
  24. 24.
    Trefethen, L.N.: Numerical Linear Algebra. SIAM, Philadelphia (1997)Google Scholar
  25. 25.
    Tsuda, K., Rätsch, G., Warmuth, M.: Matrix exponentiated gradient updates for on-line learning and Bregman projection. J. Mach. Learn. Res. 6, 995–1018 (2005)MathSciNetzbMATHGoogle Scholar
  26. 26.
    Warmuth, M., Kuzmin, D.: Online variance minimization. In: 19\(^{th}\) COLT, pp. 514–528. (2006)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  • Richard Nock
    • 1
  • Brice Magdalou
    • 1
  • Eric Briys
    • 1
  • Frank Nielsen
    • 2
  1. 1.CEREGMIA-Université Antilles-GuyaneMartiniqueFrance
  2. 2.Sony CS Labs Inc.TokyoJapan

Personalised recommendations