Inference for circular distributions and processes
One approach to defining models for circular data and processes has been to take a standard Euclidean model and to wrap it around the circle. This generates rich families of circular models but creates difficulties for inference. Using data augmentation ideas which have previously been applied to this problem in the framework of an EM algorithm, we demonstrate the power and flexibility of Markov chain Monte Carlo methods to fit such classes of models to circular data. The precision of the technique is confirmed through simulated examples, and then applications are given to multivariate and time series data of wind directions.
Unable to display preview. Download preview PDF.
- Anderson, C. W. and Nadarajah, S. (1993) Environmental factors affecting reservoir safety. In V. Barnett and K. F. Turkman (eds) Statistics for the Environment, 2, Chichester: Wiley.Google Scholar
- Breckling, J. (81 or 89) The Analysis of Directional Time Series: Applications to Wind Speed and Direction, London: Springer.Google Scholar
- Casson, E. and Coles, S. (1997) Extreme hurricane wind speeds: estimation, extrapolation and spatial smoothing. In Proceedings of the 2nd European and African Wind Engineering Conference, Genova, Italy.Google Scholar
- Cook, N. J. (1985) The Designer's Guide to Wind Loading of Building Structures, Part 1, London: Butterworths.Google Scholar
- Fisher, N. I. (1993) The Analysis of Circular Data, Cambridge: Cambridge University Press.Google Scholar
- Fisher, N. I. and Lee, A. J. (1994) Time series analysis of circular data. Journal of the Royal Statistical Society, B, 56, 327–39.Google Scholar
- Mardia, K. V. (1972) Statistics of Directional Data, London: Academic Press.Google Scholar
- Smith, A. F. M. and Roberts, G. O. (1993) Bayesian computation via the Gibbs sampler and related Markov chain Monte Carlo methods (with discussion). Journal of the Royal Statistical Society, B, 55, 3–23.Google Scholar