This article presents a Bayesian model for the clustering of non-ordered multivariate directional or circular data. The particular trait of our data is that each single observation is made up of non-ordered points on the circle. We introduce a hierarchical model that combines a symmetrization technique, projected normal distributions and a Dirichlet process. One parameter is introduced to model the non-ordered trait and another one to control the variability of the angles on the circle. An informative prior on the relative locations of the angles is also provided. The gain of the symmetrization is highlighted by a theoretical study. The parameters of the model are then inferred using a Metropolis–Hastings-within-Gibbs algorithm. Simulated datasets are analysed to study the sensitivity to hyperparameters. Then, the benefits of our approach are illustrated by clustering real data made up of the positions of five separate radiotherapy X-ray beams on a circle.
AbrahamCMolinariNServienR (2013) Unsupervised clustering of multivariate circular data. Statistics in Medicine, 32, 1376–82.
2.
BlackwellDMacQueenJB (1973) Ferguson distributions via Polya urn schemes. The Annals of Statistics, 1, 353–55.
3.
DahlDB (2003) An improved merge-split sampler for conjugate Dirichlet process mixture models (Technical Report No. 1086). Madison, WI: University of Wisconsin, 1–32.
4.
DamienPWalkerS (1999) A full Bayesian analysis of circular data using the von Mises distribution. The Canadian Journal of Statistics, 27, 291–98.
5.
EscobarMDWestM (1995) Bayesian density estimation and inference using mixtures. Journal of the American Statistical Association, 90, 577–88.
6.
FergusonTS (1973) A Bayesian analysis of some nonparametric problems. The Annals of Statistics, 1, 209–30.
7.
FritschAIckstadtK (2009) Improved criteria for clustering based on the posterior simi- larity matrix. Bayesian Analysis, 4, 367–92.
8.
GriffinJHolmesC (2010) Computational issues arising in Bayesian nonparametric hierarchical models. In Bayesian Nonpa- rametrics, edited by HjortNHolmesCMllerPWalkerSG pages 208–22. Cambridge: Cambridge University Press.
9.
Hernandez-StumpfhauserDBreidtFJvan der WoerdMJ (2017) The general projected normal distribution of arbitrary dimension: Modeling and Bayesian inference. Bayesian Analysis, 12, 113–33.
10.
HubertLArabieP (1985) Comparing part- itions. Journal of Classication, 2, 193–218.
11.
JainSNealRM (2004) A split-merge Markov chain Monte-Carlo procedure for the Dirichlet process mixture model. Journal of Computational and Graphical Statistics, 13, 158–82.
12.
Jona-LasinioGGelfandAJona-LasinioM (2012) Spatial analysis of wave direction data using wrapped Gaussian processes. The Annals of Applied Statistics, 6, 1478–98.
13.
LauJWGreenPJ (2007) Bayesian model-based clustering procedures. Journal of Computational and Graphical Statistics, 16, 526–58.
14.
MacEachernSN (1994) Estimating normal means with a conjugate style Dirichlet process prior. Communications in Statistics: Simulation and Computation, 23, 727–41.
15.
MacEachernSN (1998) Computational methods for mixture of Dirichlet process models. In Practical Nonparametric and Semipara- metric Bayesian Statistics, edited by DeyDMller andPSinhaD pages 23–44. Lecture Notes in Statistics 133. New York, NY: Springer-Verlag.
16.
MardiaKVHuguesGTaylorCCSinghH (2008) A multivariate von Mises distri- bution with applications to bioinformatics. The Canadian Journal of Statistics, 36, 99–109.
17.
MardiaKVJuppP (2009) Directional Statistics. New York, NY: John Wiley & Sons.
18.
MastrantonioGPolliceAFedeleF (2018) Distributions-oriented wind forecast verification by a hidden Markov model for multivariate circular-linear data. Stochastic Environmental Research and Risk Assessment, 32, 169–81.
19.
NealRM (2000) Markov chain sampling method for Dirichlet process mixture models. Journal of Computational and Graphical Statistics, 9, 249–65.
20.
NguyenXEppsJBaileyJ (2009) Information theoretic measures for clustering comparison: Is a correction for chance necessary? ICML’09: Proceedings of the 26th Annual International Conference on Machine Learning, Montreal, Canada, 14–18June 2009, pages 1073–80.
21.
Nuñez-AntonioGGutiérrez-PeñaE (2005) A Bayesian analysis of directional data using the projected normal distribution. Journal of Applied Statistics, 32, 995–1001.
22.
Nuñez-AntonioGGutiérrez-PeñaEEscarelaG (2011) A Bayesian regression model for circular data based on the projected normal distribution. Statistical Modelling, 11, 185–201.
23.
PresnellBMorrisonSPLittellRC (1998) Projected multivariate linear models for directional data. Journal of the American Statistical Association, 93, 1068–77.
24.
QuintanaFA (2006) A predictive view of Bayesian clustering. Journal of Statistical Planning and Inference, 136, 2407–29.
25.
RandW (1971) Objective criteria for the evalu- ation of clustering methods. Journal of the American Statistical Association, 66, 846–50.
26.
RavidranPGhoshSK (2011) Bayesian analysis of circular data using wrapped distributions. Journal of Statistical Theory and Practice, 5, 547–60.
27.
SenGuptaALahaAK (2008) A Bayesian analysis of the change-point problem for directional data. Journal of Applied Statistics, 35, 693–700.
28.
SinghHHnizdoVDemchukE (2002) Probabilistic model for two dependant circular variables. Biometrika, 89, 719–23.
29.
TehYW (2010) Dirichlet processes. In Ency- clopedia of Machine Learning. Springer.
30.
Von MisesR (1918) Über die ganzzahligkeit der atomgewicht und verwandte fragen. Physikalische Zeitschrift, 19, 490–500.
31.
WangFGelfandAE (2013) Directional data analysis under the general projected normal distribution. Statistical Methodology, 10, 113–27.
32.
WangFGelfandAE (2014) Modeling space and space-time directional data using projected Gaussian processes. Journal of the American Statistical Association, 109, 1565–80.
33.
WangFGelfandAEJona-LasinioG (2015) Joint spatio-temporal analysis of a linear and a directional variable: Space-time modeling of wave heights and wave directions in the adriatic sea. Statistica Sinica, 25, 25–9.
34.
YuanLWuQJYinFLiYShengYKelseyCRGeY (2015) Standardized beam bouquets for lung IMRT planning. Physics in Medicine & Biology, 60, 1821–43.
