Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences
Restricted accessResearch article

Gaussian processes for time-series modelling

S. Roberts

S. Roberts

Department of Engineering Science, University of Oxford, Oxford OX1 3PU, UK

[email protected]

Google Scholar

Find this author on PubMed

,
M. Osborne

M. Osborne

Department of Engineering Science, University of Oxford, Oxford OX1 3PU, UK

Google Scholar

Find this author on PubMed

,
M. Ebden

M. Ebden

Department of Engineering Science, University of Oxford, Oxford OX1 3PU, UK

Google Scholar

Find this author on PubMed

,
S. Reece

S. Reece

Department of Engineering Science, University of Oxford, Oxford OX1 3PU, UK

Google Scholar

Find this author on PubMed

,
N. Gibson

N. Gibson

Department of Astrophysics, University of Oxford, Oxford OX1 3PU, UK

Google Scholar

Find this author on PubMed

and
S. Aigrain

S. Aigrain

Department of Astrophysics, University of Oxford, Oxford OX1 3PU, UK

Google Scholar

Find this author on PubMed

Published:https://doi.org/10.1098/rsta.2011.0550

    In this paper, we offer a gentle introduction to Gaussian processes for time-series data analysis. The conceptual framework of Bayesian modelling for time-series data is discussed and the foundations of Bayesian non-parametric modelling presented for Gaussian processes. We discuss how domain knowledge influences design of the Gaussian process models and provide case examples to highlight the approaches.

    Footnotes

    One contribution of 17 to a Discussion Meeting Issue ‘Signal processing and inference for the physical sciences’.

    References

    • 1
      Rasmussen CE& Williams CKI. 2006 Gaussian processes for machine learning. Cambridge, MA: MIT Press. Google Scholar
    • 2
      Osborne MA, Rogers A, Ramchurn S, Roberts SJ& Jennings NR. 2008 Towards real-time information processing of sensor network data using computationally efficient multi-output Gaussian processes. Int. Conf. on Information Processing in Sensor Networks (IPSN 2008), April 2008, pp.109–120. IEEE Computer Society. CrossrefGoogle Scholar
    • 3
      Abrahamsen P. 1997 A review of Gaussian random fields and correlation functions. 2nd edn. Technical Report no. 917, Norwegian Computing Center, Box 114, Blindern, 0314 Oslo, Norway. Google Scholar
    • 4
      Stein ML. 2005 Space-time covariance functions. J. Am. Stat. Assoc. 100, 310–322.doi:10.1198/016214504000000854 (doi:10.1198/016214504000000854). Crossref, ISIGoogle Scholar
    • 5
      Sasena MJ. 2002 Flexibility and efficiency enhancements for constrained global design optimization with Kriging approximations. PhD thesis, University of Michigan, Ann Arbor, Michigan. Google Scholar
    • 6
      Pinheiro J& Bates D. 1996 Unconstrained parameterizations for variance–covariance matrices. Stat. Comput. 6, 289–296.doi:10.1007/BF00140873 (doi:10.1007/BF00140873). Crossref, ISIGoogle Scholar
    • 7
      Osborne M, Roberts S, Rogers A& Jennings N. In press. Real-time information processing of environmental sensor network data. Trans. Sensor Networks 9. Google Scholar
    • 8
      MacKay DJC. 1998 Introduction to Gaussian processes. Neural networks and machine learning (ed. & Bishop CM), pp. 84–92. Berlin, Germany: Springer. Google Scholar
    • 9
      Garnett R, Osborne MA& Roberts SJ. 2009 Sequential Bayesian prediction in the presence of changepoints. Proc. 26th Annual Int. Conf. Machine Learning, Montreal, June 2009, pp. 345–352. New York, NY: ACM. Google Scholar
    • 10
      Osborne M, Reece S, Rogers A, Roberts S& Garnett R. 2010 Sequential Bayesian prediction in the presence of changepoints and faults. Comp. J. 53, 1430–1446. Crossref, ISIGoogle Scholar
    • 11
      Reece S, Garnett R, Osborne MA& Roberts SJ. 2009 Anomaly detection and removal using non-stationary Gaussian processes. Technical report PARG-09-01 University of Oxford, Oxford, UK. Google Scholar
    • 12
      Reece S, Claxton C, Nicholson D& Roberts SJ. 2009 Multi-sensor fault recovery in the presence of known and unknown fault types. Proc. 12th Int. Conf. on Information Fusion (FUSION 2009), Seattle, WA, pp. 1696–1703. IEEE Computer Society. Google Scholar
    • 13
      MacKay DJC. 2002 Information theory, inference and learning algorithms. Cambridge, UK: Cambridge University Press. Google Scholar
    • 14
      Bishop C. 2006 Pattern recognition and machine learning. Berlin, Germany: Springer. Google Scholar
    • 15
      Chen J& Gupta AK. 2000 Parametric statistical change point analysis. Berlin, Germany: Birkháuser. CrossrefGoogle Scholar
    • 16
      Duda RO, Hart PE& Stork DG. 2000 Pattern classification 2nd edn. London, UK: Wiley-Interscience. Google Scholar
    • 17
      O’Hagan A. 1987 Monte Carlo is fundamentally unsound. Statistician 36, 247–249.doi:10.2307/2348519 (doi:10.2307/2348519). CrossrefGoogle Scholar
    • 18
      O’Hagan A. 1991 Bayes–Hermite quadrature. J. Stat. Plan. Inference 29, 245–260.doi:10.1016/0378-3758(91)90002-V (doi:10.1016/0378-3758(91)90002-V). Crossref, ISIGoogle Scholar
    • 19
      Rasmussen CE& Ghahramani Z. 2003 Bayesian Monte Carlo. Advances in neural information processing systems, vol. 15, pp. 489–496. Cambridge, MA: MIT Press. Google Scholar
    • 20
      Whitcher B, Guttorp P& Percival DB. 2002 Testing for homogeneity of variance in time series: long memory, wavelets and the Nile River. Water Resour. Res. 38, 10–1029.doi:10.1029/2001WR000509 (doi:10.1029/2001WR000509). Crossref, ISIGoogle Scholar
    • 21
      Ray BK& Tsay RS. 2002 Bayesian methods for change-point detection in long-range dependent processes. J. Time Series Anal. 23, 687–705.doi:10.1111/1467-9892.00286 (doi:10.1111/1467-9892.00286). Crossref, ISIGoogle Scholar
    • 22
      Adams RP& MacKay DJC. 2007 Bayesian online changepoint detection. (http://arxiv.org/abs/0710.3742). Google Scholar
    • 23
      Pont F, Aigrain S& Zucker S. 2011 A simple method to estimate radial velocity variations due to stellar activity using photometry. Mon. Not. R. Astron. Soc. 419, 3147. ISIGoogle Scholar
    • 24
      Henry GW& Winn JN. 2008 The rotation period of the planet-hosting star HD 189733. Astron. J. 135, 68.doi:10.1088/0004-6256/135/1/68 (doi:10.1088/0004-6256/135/1/68). Crossref, ISIGoogle Scholar
    • 25
      Gibson NP, Aigrain S, Roberts S, Evans T, Osborne M& Pont F. In press. A Gaussian process framework for modelling instrumental systematics: application to transmission spectroscopy. Mon. Not. R. Astron. Soc.. Google Scholar
    • 26
      Luengo D, Lawrence N& Álvarez M. 2009 Latent force models. Proc. 12th Int. Workshop on Artificial Intelligence and Statistics, Clearwater Beach, FL, 16–18 April 2009, pp. 9–16. JMLR Workshop and Conference Proceedings, vol. 5. MIT Press. Google Scholar