Abstract
A Bayesian approach to modelling binary data on a regular lattice is introduced. The method uses a hierarchical model where the observed data is the sign of a hidden conditional autoregressive Gaussian process. This approach essentially extends the familiar probit model to dependent data. Markov chain Monte Carlo simulations are used on real and simulated data to estimate the posterior distribution of the spatial dependency parameters and the method is shown to work well. The method can be straightforwardly extended to regression models.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Albert, P. S. and McShane, L. M. (1995) Generalized estimating equations approach to spatially correlated data: applications to the analysis of neuroimaging data. Biometrics, 51, 627–38.
Besag, J. E. (1975) Statistical analysis of non-lattice data. The Statistician, 24, 179–95.
Besag, J. E., Green, P. J., Higdon, D. and Mengersen, K. (1995) Bayesian computation and stochastic systems. Statistical Science, 10, 3–66.
Besag, J. E. and Moran, P. A. P. (1975) On the estimation and testing of spatial interaction in Gaussian lattice processes. Biometrika, 62, 555–62.
Best, N. G., Cowles, M. K. and Vines, S. K. (1995) CODA Manual version 0.30. MRC Biostatistics Unit, Cambridge, UK.
Cressie, N. A. C. (1993). Statistics for spatial data, Revised Edi-tion, Wiley, New York.
Diggle, P. J., Tawn, J. A. and Moyeed, R. A. (1998) Model-based geostatistics (with discussion). Appl. Statist., 47, 299–350.
Gelman, A. and Meng, X.-L. (1998) Simulating Normalizing Constants: From Importance Sampling to Bridge Sampling to Path Sampling. Statist. Sci., 13, 163–85.
Geyer, C. J. (1991) Markov chain Monte Carlo maximum likeli-hood. Computer Science and Statistics: Proc. 23rd Symp. Interface, 156–63.
Geyer, C. J. and Thompson, E. A. (1992) Constrained Monte Carlo maximum likelihood for dependent data. J. Roy. Statist. Soc., B, 657–99.
Gilks, W. R., Richardson, S. and Spiegelhalter, D. J. (1996) Markov Chain Monte Carlo In Practice, London: Chapman and Hall.
Harvey, A. C. and Durbin, J. (1986) The effects of seat belt leg-islation on British road casualties: a case study in structural time series modelling. J. Roy. Statist. Sco., A, 149, 187–227.
Hay, J. and Pettitt, A. N. (1998) Bayesian analysis of a Time Series of counts. Submitted to Biometrics.
Heidelberger, P. and Welch, P. (1983) Simulation run length control in the presence of an initial transient. Operations Research, 7, 493–97.
Liang, K.-Y. and Zeger, S. L. (1986) Longitudinal data analysis using generalized linear models Biometrika, 87, 561–74.
Low Choy, S. (1998) Bayesian hierarchical modelling of two-di-mensional lattice data: a biogeographical case study. Queensland University of Technology PhD thesis.
Mengersen, K. L. and Robert, C. P. (1998) MCMC conver-gence diagnostics: a ``revieWWW'', Proceedings of Sixth Valencia International Meeting on Bayesian Statistics, 399–432.
Moran, P. A. P. (1973) A Gaussian Markovian process on a square lattice. J. Appl. Prob., 10, 54–62.
Pettitt, A. N. and Low Choy, S. (1999) Bivariate binary data with missing values: analysis of a field experiment to investigate chemical attractants of wild dogs. Submitted to J. Agric. Biol. Env. Stats.
Stein, M. (1992). Prediction and inference for truncated spatial data. J. Comput. Graph. Statist., 1, 91–110.
Rights and permissions
About this article
Cite this article
Weir, I.S., Pettitt, A.N. Spatial modelling for binary data using␣a␣hidden conditional autoregressive Gaussian process: a multivariate extension of the probit model. Statistics and Computing 9, 77–86 (1999). https://doi.org/10.1023/A:1008822429149
Issue Date:
DOI: https://doi.org/10.1023/A:1008822429149