Bayesian Analysis

Dynamic Chain Graph Models for Time Series Network Data

Osvaldo Anacleto and Catriona Queen

Full-text: Open access

Abstract

This paper introduces a new class of Bayesian dynamic models for inference and forecasting in high-dimensional time series observed on networks. The new model, called the dynamic chain graph model, is suitable for multivariate time series which exhibit symmetries within subsets of series and a causal drive mechanism between these subsets. The model can accommodate high-dimensional, non-linear and non-normal time series and enables local and parallel computation by decomposing the multivariate problem into separate, simpler sub-problems of lower dimensions. The advantages of the new model are illustrated by forecasting traffic network flows and also modelling gene expression data from transcriptional networks.

Article information

Source
Bayesian Anal., Volume 12, Number 2 (2017), 491-509.

Dates
First available in Project Euclid: 17 June 2016

Permanent link to this document
https://projecteuclid.org/euclid.ba/1466165926

Digital Object Identifier
doi:10.1214/16-BA1010

Mathematical Reviews number (MathSciNet)
MR3620742

Zentralblatt MATH identifier
1384.62285

Keywords
chain graph multiregression dynamic model network traffic flow forecasting gene expression networks network data time series

Rights
Creative Commons Attribution 4.0 International License.

Citation

Anacleto, Osvaldo; Queen, Catriona. Dynamic Chain Graph Models for Time Series Network Data. Bayesian Anal. 12 (2017), no. 2, 491--509. doi:10.1214/16-BA1010. https://projecteuclid.org/euclid.ba/1466165926


Export citation

References

  • Abegaz, F. and Wit, E. (2013). “Sparse time series chain graphical models for reconstructing genetic networks.” Biostatistics, 14(3): 586–599.
  • Anacleto, O. (2012). “Bayesian dynamic graphical models for high-dimensional flow forecasting in road traffic networks.” Ph.D. thesis, The Open University.
  • Anacleto, O. and Queen, C. (2016). “Supplementary material for paper: Dynamic chain graph models for time series network data.” Bayesian Analysis.
  • Anacleto, O., Queen, C., and Albers, C. J. (2013a). “Forecasting multivariate road traffic flows using Bayesian dynamic graphical models, splines and other traffic variables.” Australian & New Zealand Journal of Statistics, 55(2): 69–86.
  • Anacleto, O., Queen, C., and Albers, C. J. (2013b). “Multivariate forecasting of road traffic flows in the presence of heteroscedasticity and measurement errors.” Journal of the Royal Statistical Society: Series C (Applied Statistics), 62(2): 251–270.
  • Bar-Joseph, Z., Gitter, A., and Simon, I. (2012). “Studying and modelling dynamic biological processes using time-series gene expression data.” Nature Reviews Genetics, 13(8): 552–564.
  • Cantone, I., Marucci, L., Iorio, F., Ricci, M. A., Belcastro, V., Bansal, M., Santini, S., Di Bernardo, M., Di Bernardo, D., and Cosma, M. P. (2009). “A yeast synthetic network for in vivo assessment of reverse-engineering and modeling approaches.” Cell, 137(1): 172–181.
  • Carvalho, C. M. and West, M. (2007). “Dynamic matrix-variate graphical models.” Bayesian Analysis, 2(1): 69–97.
  • Costa, L., Smith, J., Nichols, T., Cussens, J., Duff, E. P., and Makin, T. R. (2015). “Searching multiregression dynamic models of resting-state fMRI networks using integer programming.” Bayesian Analysis, 10(2): 441–478.
  • Cox, D. R. and Wermuth, N. (1996). Multivariate Dependencies: Models, Analysis and Interpretation, volume 67. CRC Press.
  • Cressie, N. and Wikle, C. K. (2011). Statistics for Spatio-Temporal Data. John Wiley & Sons.
  • Dawid, A. P. (1981). “Some matrix-variate distribution theory: notational considerations and a Bayesian application.” Biometrika, 68(1): 265–274.
  • Gruber, L. F. and West, M. (2015). “GPU-accelerated Bayesian learning in simultaneous graphical dynamic linear models.” Bayesian Analysis, 11(1): 125–149.
  • Husmeier, D., Werhli, A. V., and Grzegorczyk, M. (2011). “Advanced Applications of Bayesian Networks in Systems Biology.” In Stumpf, M. P. H., Balding, D. J., and Girolami, M. (eds.), Handbook of Statistical Systems Biology, 270–289. Wiley Online Library.
  • Kolaczyk, E. D. (2009). Statistical Analysis of Network Data: Methods and Models. Springer, New York.
  • Lauritzen, S. L. (1996). Graphical Models. Oxford University Press.
  • Lèbre, S. (2009). “Inferring dynamic genetic networks with low order independencies.” Statistical Applications in Genetics and Molecular Biology, 8(1): 1–38.
  • Li, B. (2009). “A non-Gaussian Kalman filter with application to the estimation of vehicular speed.” Technometrics, 51(2): 162–172.
  • McCarter, C. and Kim, S. (2014). “On Sparse Gaussian Chain Graph Models.” In Advances in Neural Information Processing Systems, 3212–3220.
  • Mohammadi, A. and Wit, E. C. (2015). “Bayesian structure learning in sparse Gaussian graphical models.” Bayesian Analysis, 10(1): 109–138.
  • Nascimento, F. F., Gamerman, D., and Lopes, H. F. (2015). “Time-varying extreme pattern with dynamic models.” TEST, 1–19.
  • Oates, C., Smith, J., Mukherjee, S., and Cussens, J. (2015a). “Exact estimation of multiple directed acyclic graphs.” Statistics and Computing, 1–15. http://dx.doi.org/10.1007/s11222-015-9570-9
  • Oates, C. J., Costa, L., and Nichols, T. E. (2015b). “Toward a multisubject analysis of neural connectivity.” Neural Computation, 27: 151–170.
  • Opgen-Rhein, R. and Strimmer, K. (2007). “From correlation to causation networks: a simple approximate learning algorithm and its application to high-dimensional plant gene expression data.” BMC Systems Biology, 1(1): 37.
  • Prado, R. and West, M. (2010). Time Series: Modeling, Computation, and Inference. CRC Press.
  • Queen, C. and Smith, J. (1992). “Dynamic graphical models.” Bayesian Statistics, 4: 741–751.
  • Queen, C. M. (1994). “Using the multiregression dynamic model to forecast brand sales in a competitive product market.” The Statistician, 87–98.
  • Queen, C. M. and Albers, C. J. (2009). “Intervention and causality: forecasting traffic flows using a dynamic Bayesian network.” Journal of the American Statistical Association, 104(486): 669–681.
  • Queen, C. M. and Smith, J. Q. (1993). “Multiregression dynamic models.” Journal of the Royal Statistical Society. Series B (Methodological), 849–870.
  • Queen, C. M., Wright, B. J., and Albers, C. J. (2007). “Eliciting a directed acyclic graph for a multivariate time series of vehicle counts in a traffic network.” Australian & New Zealand Journal of Statistics, 49(3): 221–239.
  • Queen, C. M., Wright, B. J., and Albers, C. J. (2008). “Forecast covariances in the linear multiregression dynamic model.” Journal of Forecasting, 27(2): 175–191.
  • Quintana, J. M. and West, M. (1987). “An analysis of international exchange rates using multivariate DLM’s.” The Statistician, 275–281.
  • Quirós, A., Wilson, S. P., Diez, R. M., Solana, A. B., and Tamames, J. A. H. (2015). “Brain activity detection by estimating the signal-to-noise ratio of fMRI time series using dynamic linear models.” Digital Signal Processing, 47: 205–211.
  • Scutari, M. (2013). “On the prior and posterior distributions used in graphical modelling (with discussion).” Bayesian Analysis, 8(3): 505–532.
  • Smith, S. M., Fulton, D. C., Chia, T., Thorneycroft, D., Chapple, A., Dunstan, H., Hylton, C., Zeeman, S. C., and Smith, A. M. (2004). “Diurnal changes in the transcriptome encoding enzymes of starch metabolism provide evidence for both transcriptional and posttranscriptional regulation of starch metabolism in Arabidopsis leaves.” Plant Physiology, 136(1): 2687–2699.
  • Stein, T., Morris, J. S., Davies, C. R., Weber-Hall, S. J., Duffy, M.-A., Heath, V. J., Bell, A. K., Ferrier, R. K., Sandilands, G. P., and Gusterson, B. A. (2004). “Involution of the mouse mammary gland is associated with an immune cascade and an acute-phase response, involving LBP, CD14 and STAT3.” Breast Cancer Research, 6(2): R75–91.
  • Velozo, P. L., Alves, M. B., and Schmidt, A. M. (2014). “Modelling categorized levels of precipitation.” Brazilian Journal of Probability and Statistics, 28(2): 190–208.
  • Wang, H. (2010). “Sparse seemingly unrelated regression modelling: Applications in finance and econometrics.” Computational Statistics & Data Analysis, 54(11): 2866–2877.
  • Wang, H. (2015). “Scaling it up: Stochastic search structure learning in graphical models.” Bayesian Analysis, 10(2): 351–377.
  • Wang, H., Reeson, C., and Carvalho, C. M. (2011). “Dynamic financial index models: Modeling conditional dependencies via graphs.” Bayesian Analysis, 6(4): 639–664.
  • Wermuth, N. and Lauritzen, S. L. (1990). “On substantive research hypotheses, conditional independence graphs and graphical chain models.” Journal of the Royal Statistical Society. Series B (Methodological), 21–50.
  • West, M. and Harrison, J. (1997). Bayesian Forecasting and Dynamic Models (2nd ed.). New York, NY, USA: Springer-Verlag New York, Inc.
  • Xiao, S., Kottas, A., and Sansó, B. (2015). “Modeling for seasonal marked point processes: An analysis of evolving hurricane occurrences.” The Annals of Applied Statistics, 9(1): 353–382.
  • Zhao, Z. Y. (2015). “Bayesian Multiregression Dynamic Models with Applications in Finance and Business.” Ph.D. thesis, Duke University.
  • Zhao, Z. Y., Xie, M., and West, M. (2015). “Dynamic dependence networks: Financial time series forecasting & portfolio decisions.” Applied Stochastic Models in Business and Industry (to appear).

Supplemental materials

  • Supplementary material for paper: Dynamic chain graph models for time series network data. Supplementary material available online includes the theorem for which Corollary 1 is a consequence, together with the proofs of that theorem and Corollary 1. It also includes the description and results of the application of the DCGM to two gene expression datasets, as mentioned in Section 5.