The Variational Bayesian Approach to Fitting Mixture Models to Circular Wave Direction Data

Burton Wu Queensland University of Technology, Brisbane, Queensland, Australia

Search for other papers by Burton Wu in
Current site
Google Scholar
PubMed
Close
,
Clare A. McGrory The University of Queensland, Brisbane, Queensland, Australia

Search for other papers by Clare A. McGrory in
Current site
Google Scholar
PubMed
Close
, and
Anthony N. Pettitt Queensland University of Technology, Brisbane, Queensland, Australia

Search for other papers by Anthony N. Pettitt in
Current site
Google Scholar
PubMed
Close
Restricted access

Abstract

The emerging variational Bayesian (VB) technique for approximate Bayesian statistical inference is a non-simulation-based and time-efficient approach. It provides a useful, practical alternative to other Bayesian statistical approaches such as Markov chain Monte Carlo–based techniques, particularly for applications involving large datasets. This article reviews the increasingly popular VB statistical approach and illustrates how it can be used to fit Gaussian mixture models to circular wave direction data. This is done by taking the straightforward approach of padding the data; this method involves adding a repeat of a complete cycle of the data to the existing dataset to obtain a dataset on the real line. The padded dataset can then be analyzed using the standard VB technique. This results in a practical, efficient approach that is also appropriate for modeling other types of circular, or directional, data such as wind direction.

Corresponding author address: Clare A. McGrory, Centre for Applications in Natural Resource Mathematics, School of Mathematics, The University of Queensland, St. Lucia, Brisbane, QLD 4072, Australia. E-mail: c.mcgrory@uq.edu.au

Abstract

The emerging variational Bayesian (VB) technique for approximate Bayesian statistical inference is a non-simulation-based and time-efficient approach. It provides a useful, practical alternative to other Bayesian statistical approaches such as Markov chain Monte Carlo–based techniques, particularly for applications involving large datasets. This article reviews the increasingly popular VB statistical approach and illustrates how it can be used to fit Gaussian mixture models to circular wave direction data. This is done by taking the straightforward approach of padding the data; this method involves adding a repeat of a complete cycle of the data to the existing dataset to obtain a dataset on the real line. The padded dataset can then be analyzed using the standard VB technique. This results in a practical, efficient approach that is also appropriate for modeling other types of circular, or directional, data such as wind direction.

Corresponding author address: Clare A. McGrory, Centre for Applications in Natural Resource Mathematics, School of Mathematics, The University of Queensland, St. Lucia, Brisbane, QLD 4072, Australia. E-mail: c.mcgrory@uq.edu.au
Save
  • Attias, H., 1999: Inferring parameters and structure of latent variable models by variational Bayes. Proc. 15th Conf. on Uncertainty in Artificial Intelligence, Stockholm, Sweden, Association for Uncertainty in Artificial Intelligence, 21–30.

  • Bishop, C., 2006: Pattern Recognition and Machine Learning. Springer, 738 pp.

  • Celeux, G., M. Hurn, and C. P. Robert, 2000: Computational and inferential difficulties with mixture posterior distributions. J. Amer. Stat. Assoc., 95, 957970.

    • Search Google Scholar
    • Export Citation
  • Corduneanu, A., and C. M. Bishop, 2001: Variational Bayesian model selection for mixture distributions. Proc. Eighth Int. Conf. on Artificial Intelligence and Statistics, Key West, FL, Society for Artificial Intelligence and Statistics, 27–34.

  • Farrugia, P., J. Borg, and A. Micallef, 2009: On the algorithms used to compute the standard deviation of wind direction. J. Appl. Meteor. Climatol., 48, 21442151.

    • Search Google Scholar
    • Export Citation
  • Gelman, A., J. Carlin, H. Stern, and D. Rubin, 2003: Bayesian Data Analysis. 2nd ed. Texts in Statistical Science, Chapman and Hall, 668 pp.

  • Goodwin, I., 2005: A mid-shelf, mean wave direction climatology for southeastern Australia, and its relationship to the El Niño–Southern Oscillation since 1878 A.D. Int. J. Climatol., 25, 17151729.

    • Search Google Scholar
    • Export Citation
  • Goodwin, I., R. Freeman, and K. Blackmore, 2010: Decadal wave climate variability and implications for interpreting New South Wales coastal behaviour. Proc. Australian Wind Waves Research Science Symp., Gold Coast, QLD, Australia, Centre for Australian Weather and Climate Research, 58–61.

  • Jammalamadaka, S. R., and A. Sengupta, 2001: Topics in Circular Statistics. World Scientific, 322 pp.

  • Jasra, A., C. C. Holmes, and D. A. Stephens, 2005: Markov chain Monte Carlo methods and the label switching problem in Bayesian mixture modeling. Stat. Sci., 20, 5067.

    • Search Google Scholar
    • Export Citation
  • Komrek, A., 2009: A new R package for Bayesian estimation of multivariate normal mixtures allowing for selection of the number of components and interval-censored data. Comput. Stat. Data Anal., 53, 39323947.

    • Search Google Scholar
    • Export Citation
  • Mackay, D. J. C., 2003: Information Theory, Inference, and Learning Algorithms. Cambridge University Press, 628 pp.

  • Mahrt, L., 2011: Surface wind direction variability. J. Appl. Meteor. Climatol., 50, 144152.

  • Mardia, K. V., and P. E. Jupp, 2000: Directional Statistics. 2nd ed. John Wiley and Sons, 429 pp.

  • McGrory, C. A., and D. M. Titterington, 2007: Variational approximations in Bayesian model selection for finite mixture distributions. Comput. Stat. Data Anal., 51, 53525367.

    • Search Google Scholar
    • Export Citation
  • McLachlan, G. J., and D. Peel, 2000: Finite Mixture Models. John Wiley and Sons, 419 pp.

  • McVinish, R., and K. Mengersen, 2008: Semiparametric Bayesian circular statistics. Comput. Stat. Data Anal., 52, 47224730.

  • Richardson, S., and P. J. Green, 1997: On Bayesian analysis of mixtures with an unknown number of components (with discussion). J. Roy. Stat. Soc., 59B, 731792.

    • Search Google Scholar
    • Export Citation
  • Shand, T., and Coauthors, 2010: NSW coastal storms and extreme waves. Proc. 19th NSW Coastal Conf., Batemans Bay, NSW, Australia, Australian Coastal Society, 14 pp. [Available online at http://www.coastalconference.com/2010/papers2010/Tom%20Shand%20full%20paper.pdf.]

  • Uppala, S. M., and Coauthors, 2005: The ERA-40 Re-Analysis. Quart. J. Roy. Meteor. Soc., 131, 29613012.

  • Weber, R., 1997: Estimators for the standard deviation of horizontal wind direction. J. Appl. Meteor. Climatol., 36, 14031415.

  • Wu, B., C. A. McGrory, and A. N. Pettitt, 2012: A new variational Bayesian algorithm with application to human mobility pattern modeling. Stat. Comput., 22, 185203.

    • Search Google Scholar
    • Export Citation
All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 1431 1143 33
PDF Downloads 293 72 9