Skip to main content
Log in

Modeling correlated samples via sparsematrix Gaussian graphical models

  • Published:
Journal of Zhejiang University SCIENCE C Aims and scope Submit manuscript

Abstract

A new procedure of learning in Gaussian graphical models is proposed under the assumption that samples are possibly dependent. This assumption, which is pragmatically applied in various areas of multivariate analysis ranging from bioinformatics to finance, makes standard Gaussian graphical models (GGMs) unsuitable. We demonstrate that the advantage of modeling dependence among samples is that the true discovery rate and positive predictive value are improved substantially than if standard GGMs are applied and the dependence among samples is ignored. The new method, called matrix-variate Gaussian graphical models (MGGMs), involves simultaneously modeling variable and sample dependencies with the matrix-normal distribution. The computation is carried out using a Markov chain Monte Carlo (MCMC) sampling scheme for graphical model determination and parameter estimation. Simulation studies and two real-world examples in biology and finance further illustrate the benefits of the new models.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

References

  • Allen, G.I., Tibshirani, R., 2010. Transposable regularized covariance models with an application to missing data imputation. Ann. Appl. Stat., 4(2):764–790. [doi:10.1214/09-AOAS314]

    Article  MathSciNet  MATH  Google Scholar 

  • Allen, G.I., Tibshirani, R., 2012. Inference with transposable data: modelling the effects of row and column correlations. J. R. Stat. Soc. B, 74(4):721–743. [doi:10.1111/j.1467-9868.2011.01027.x]

    Article  MathSciNet  Google Scholar 

  • Atay-Kayis, A., Massam, H., 2005. The marginal likelihood for decomposable and non-decomposable graphical Gaussian models. Biometrika, 92(2):317–335. [doi:10.1093/biomet/92.2.317]

    Article  MathSciNet  MATH  Google Scholar 

  • Carvalho, C.M., West, M., 2007. Dynamic matrix-variate graphical models. Bayes. Anal., 2(1):69–98. [doi:10. 1214/07-BA204]

    Article  MathSciNet  Google Scholar 

  • Castelo, R., Roverato, A., 2006. A robust procedure for Gaussian graphical model search from microarray data with p larger than n. J. Mach. Learn. Res., 7:2621–2650.

    MathSciNet  MATH  Google Scholar 

  • Dawid, A.P., Lauritzen, S.L., 1993. Hyper-Markov laws in the statistical analysis of decomposable graphical models. Ann. Stat., 21(3):1272–1317. [doi:10.1214/aos/1176349260]

    Article  MathSciNet  MATH  Google Scholar 

  • Dellaportas, P., Giudici, P., Roberts, G., 2003. Bayesian inference for nondecomposable graphical Gaussian models. Sankhya Ser. A, 65:43–55.

    MathSciNet  Google Scholar 

  • Efron, B., 2009. Are a set of microarrays independent of each other? Ann. Appl. Stat., 3(3):922–942.

    Article  MathSciNet  MATH  Google Scholar 

  • Giudici, P., Green, P.J., 1999. Decomposable graphical Gaussian model determination. Biometrika, 86(4):785–801. [doi:10.1093/biomet/86.4.785]

    Article  MathSciNet  MATH  Google Scholar 

  • Jones, B., Carvalho, C., Dobra, A., Hans, C., Carter, C., West, M., 2005. Experiments in stochastic computation for high-dimensional graphical models. Stat. Sci., 20(4):388–400. [doi:10.1214/088342305000000304]

    Article  MathSciNet  MATH  Google Scholar 

  • Kociecki, A., Rubaszek, M., Ca’Zorzi, M., 2012. Bayesian Analysis of Recursive SVAR Models with Overidentifying Restrictions. Eurosystem Working Paper Series, No. 1492.

  • Lauritzen, S.L., 1996. Graphical Models. Clarendon Press, Oxford.

    Google Scholar 

  • Pástor, L., Stambaugh, R.F., 2002. Mutual fund performance and seemingly unrelated assets. J. Financ. Econ., 63(3):315–349. [doi:10.1016/S0304-405X(02)00064-8]

    Article  Google Scholar 

  • Petersen, M.A., 2009. Estimating standard errors in finance panel data sets: comparing approaches. Rev. Financ. Stud., 22(1):435–480. [doi:10.1093/rfs/hhn053]

    Article  Google Scholar 

  • Rajaratnam, B., Massam, H., Carvalho, C.M., 2008. Flexible covariance estimation in graphical Gaussian models. Ann. Stat., 36(6):2818–2849. [doi:10.1214/08-AOS619]

    Article  MathSciNet  MATH  Google Scholar 

  • Roverato, A., 2002. Hyper-inverse Wishart distribution for non-decomposable graphs and its application to Bayesian inference for Gaussian graphical models. Scand. J. Stat., 29(3):391–411. [doi:10.1111/1467-9469. 00297]

    Article  MathSciNet  MATH  Google Scholar 

  • Sharpe, W.F., 1964. Capital asset prices: a theory of market equilibrium under conditions of risk. J. Finance, 19(3):425–442. [doi:10.2307/2977928]

    MathSciNet  Google Scholar 

  • Wang, H., 2010. Sparse seemingly unrelated regression modelling: applications in finance and econometrics. Comput. Stat. Data Anal., 54(11):2866–2877. [doi:10.1016/j.csda.2010.03.028]

    Article  Google Scholar 

  • Wang, H., Li, S.Z.Z., 2012. Efficient Gaussian graphical model determination under G-Wishart prior distributions. Electron. J. Stat., 6:168–198. [doi:10.1214/12-EJS669]

    Article  MathSciNet  Google Scholar 

  • Wang, H., West, M., 2009. Bayesian analysis of matrix normal graphical models. Biometrika, 96(4):821–834. [doi:10.1093/biomet/asp049]

    Article  MathSciNet  MATH  Google Scholar 

  • Wang, H., Reeson, C., Carvalho, C.M., 2011. Dynamic financial index models: modeling conditional dependencies via graphs. Bayes. Anal., 6:639–664. [doi:10.1214/11-BA624]

    Article  MathSciNet  Google Scholar 

  • Wooldridge, J.M., 2001. Econometric Analysis of Cross Section and Panel Data. The MIT Press.

  • Zhang, G., Ferrari, S., Qian, M., 2009. An information roadmap method for robotic sensor path planning. J. Intell. Robot. Syst., 56(1–2):69–98. [doi:10.1007/s10846-009-9318-x]

    Article  MATH  Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Hao Wang.

Rights and permissions

Reprints and permissions

About this article

Cite this article

He, Yz., Chen, X. & Wang, H. Modeling correlated samples via sparsematrix Gaussian graphical models. J. Zhejiang Univ. - Sci. C 14, 107–117 (2013). https://doi.org/10.1631/jzus.C1200316

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1631/jzus.C1200316

Key words

CLC number

Navigation