Advertisement

A Fast Metropolis-Hastings Method for Generating Random Correlation Matrices

  • Irene CórdobaEmail author
  • Gherardo Varando
  • Concha Bielza
  • Pedro Larrañaga
Conference paper
  • 1.2k Downloads
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11314)

Abstract

We propose a novel Metropolis-Hastings algorithm to sample uniformly from the space of correlation matrices. Existing methods in the literature are based on elaborated representations of a correlation matrix, or on complex parametrizations of it. By contrast, our method is intuitive and simple, based the classical Cholesky factorization of a positive definite matrix and Markov chain Monte Carlo theory. We perform a detailed convergence analysis of the resulting Markov chain, and show how it benefits from fast convergence, both theoretically and empirically. Furthermore, in numerical experiments our algorithm is shown to be significantly faster than the current alternative approaches, thanks to its simple yet principled approach.

Keywords

Correlation matrices Random sampling Metroplis-Hastings 

Notes

Acknowledgements

This work has been partially supported by the Spanish Ministry of Economy, Industry and Competitiveness through the Cajal Blue Brain (C080020-09; the Spanish partner of the EPFL Blue Brain initiative) and TIN2016-79684-P projects; by the Regional Government of Madrid through the S2013/ICE-2845-CASI-CAM-CM project; and by Fundación BBVA grants to Scientific Research Teams in Big Data 2016. I. Córdoba has been supported by the predoctoral grant FPU15/03797 from the Spanish Ministry of Education, Culture and Sports. G. Varando has been partially supported by research grant 13358 from VILLUM FONDEN.

References

  1. 1.
    Marsaglia, G., Olkin, I.: Generating correlation matrices. SIAM J. Sci. Stat. Comput. 5(2), 470–475 (1984)MathSciNetCrossRefGoogle Scholar
  2. 2.
    Holmes, R.: On random correlation matrices. SIAM J. Matrix Anal. Appl. 12(2), 239–272 (1991)MathSciNetCrossRefGoogle Scholar
  3. 3.
    Fallat, S., Lauritzen, S., Sadeghi, K., Uhler, C., Wermuth, N., Zwiernik, P.: Total positivity in Markov structures. Ann. Stat. 45(3), 1152–1184 (2017)MathSciNetCrossRefGoogle Scholar
  4. 4.
    Pourahmadi, M., Wang, X.: Distribution of random correlation matrices: hyperspherical parameterization of the Cholesky factor. Stat. Prob. Lett. 106, 5–12 (2015)MathSciNetCrossRefGoogle Scholar
  5. 5.
    Lewandowski, D., Kurowicka, D., Joe, H.: Generating random correlation matrices based on vines and extended onion method. J. Multivar. Anal. 100(9), 1989–2001 (2009)MathSciNetCrossRefGoogle Scholar
  6. 6.
    Laurent, M., Poljak, S.: On the facial structure of the set of correlation matrices. SIAM J. Matrix Anal. Appl. 17(3), 530–547 (1996)MathSciNetCrossRefGoogle Scholar
  7. 7.
    Diaconis, P., Holmes, S., Shahshahani, M.: Sampling from a Manifold, Collections, vol. 10, pp. 102–125. Institute of Mathematical Statistics (2013)Google Scholar
  8. 8.
    Eaton, M.L.: Multivariate Statistics: A Vector Space Approach. Wiley, Hoboken (1983)zbMATHGoogle Scholar
  9. 9.
    Mardia, K., Jupp, P.: Directional Statistics. Wiley, Hoboken (1999)CrossRefGoogle Scholar
  10. 10.
    Pukkila, T.M., Rao, C.R.: Pattern recognition based on scale invariant discriminant functions. Inf. Sci. 45(3), 379–389 (1988)MathSciNetCrossRefGoogle Scholar
  11. 11.
    Robert, C., Casella, G.: Monte Carlo Statistical Methods. Springer, New York (2004).  https://doi.org/10.1007/978-1-4757-4145-2CrossRefzbMATHGoogle Scholar
  12. 12.
    R Core Team: R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna (2018)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2018

Authors and Affiliations

  1. 1.Department of Artificial IntelligenceUniversidad Politécnica de MadridMadridSpain
  2. 2.Department of Mathematical SciencesUniversity of CopenhagenCopenhagenDenmark

Personalised recommendations