Abstract
This paper introduces a new family of local density separations for assessing robustness of finite-dimensional Bayesian posterior inferences with respect to their priors. Unlike for their global equivalents, under these novel separations posterior robustness is recovered even when the functioning posterior converges to a defective distribution, irrespectively of whether the prior densities are grossly misspecified and of the form and the validity of the assumed data sampling distribution. For exponential family models, the local density separations are shown to form the basis of a weak topology closely linked to the Euclidean metric on the natural parameters. In general, the local separations are shown to measure relative roughness of the prior distribution with respect to its corresponding posterior and provide explicit bounds for the total variation distance between an approximating posterior density to a genuine posterior. We illustrate the application of these bounds for assessing robustness of the posterior inferences for a dynamic time series model of blood glucose concentration in diabetes mellitus patients with respect to alternative prior specifications.
Similar content being viewed by others
References
Cadre B. (2004) Asymptotic Bayesian robustness in Bayesian decision theory. The Annals of Statistics 32: 1341–1366
Andrade J.A.A., O’Hagan A. (2006) Bayesian robustness modelling using regularly varying distributions. Bayesian Analysis 1: 169–188
Berger J. O. (1992) Recent methodological advances in robust Bayesian inference (with discussion). In: Bernardo J.M., Berger J.O., Dawid A.P., Smith A.F.M. (eds) Bayesian Statistics 4, Proceedings of the Fourth Valencia International Meeting. Clarendon Press, Oxford, pp 495–496 [In discussion of Wasserman, L.(1992b)]
Berger J.O., Wolpert R.L. (1984) The likelihood principle. In: Gupta S.S. (eds) IMS Lecture Notes (Vol. 6.) Hayward, CA: IMS
Bernardo J.M., Smith A.F.M. (1996) Bayesian theory. Wiley, Chichester
Copas J., Eguchi S. (2010) Likelihood for statistically equivalent models. Journal of the Royal Statistical Society B 72: 193–217
Daneseshkhah, A. (2004). Estimation in causal graphical models. PhD Thesis, University of Warwick, UK.
Dawid A. P. (1973) Posterior expectations for large observations. Biometrika 60: 664–667
De Robertis, L. (1978). The use of partial prior knowledge in Bayesian inference. PhD Thesis, Yale University, CT, USA.
Fernandez C., Osiewalski J., Steel M. (1996) Classical and Bayesian inference robustness in multivariate regression models. Journal of the American Statistical Association 92: 1434–1444
Gelfand A.E., Smith A.F.M. (1990) Sampling-based approaches to calculating marginal densities. Journal of the American Statistical Association 85: 398–409
Geyer C.J., Thompson E.A. (1995) Annealing Markov chain Monte Carlo with applications to ancestral inference. Journal of the American Statistical Association 90: 909–920
Ghosh J.K., Ramamoorthi R.V. (2003) Bayesian nonparametrics. Springer Series in Statistics. Springer, New York
Gustafson P. (1996) Aspects of Bayesian robustness in hierarchical models. In: Berger J.O., Betro B., Moreno E., Pericchi L.R., Ruggeri F., Salinetti G., Wasserman L. (eds) IMS Lecture Notes (Vol. 29). Hayward, CA: IMS, pp 81–100
Gustafson P., Bose S. (1996) Aspects of Bayesian robustness in hierarchical models. In: Berger J.O., Betro B., Moreno E., Pericchi L.R., Ruggeri F., Salinetti G., Wasserman L. (eds) IMS Lecture Notes (Vol. 29). Hayward, CA: IMS, pp 63–80
Gustafson P., Wasserman L. (1995) Local sensitivity diagnostics for Bayesian inference. The Annals of Statistics 23: 2153–2167
Huber P.J. (1997) Robustness: where are we now?. In: Dodge Y. (eds) IMS Lecture Notes (Vol. 31). Hayward, CA: IMS, pp 487–498
Kadane J., Srinivasan C., Salinetti G. (1996) Bayesian robustness and stability. In: Berger J.O., Betro B., Moreno E., Pericchi L.R., Ruggeri F., Salinetti G., Wasserman L. (eds) IMS Lecture Notes (Vol. 29). Hayward, CA IMS, pp 81–100
Kirkpatrick S., Gelatt C.D., Vecchi M.P. (1983) Optimization by Simulated Annealing. Science 220: 671–680
Martin J., Rios Insua D., Ruggeri F. (1998) Issues in Bayesian loss robustness. The Indian Journal of Statistics (A) 60: 405–416
Monhor D. (2007) A Chebyshev inequality for multivariate normal distributions. Probability in the Engineering and Informational Sciences 21: 289–300
Moran P. A. P. (1968) An introduction to probability theory. Oxford University Press, Oxford
O’Hagan A. (1979) On outlier rejection phenomena in Bayesian inference. Journal of the Royal Statistical Society B 41: 358–367
O’Hagan A. (2006) Eliciting expert beliefs in substantial practical applications. The Statistician 47: 21–35
O’Hagan A., Forster J. (2004) Bayesian inference. In Kendall’s Advanced Theory of Statistics (Vol. 2B). Arnold, London
Peterka V. (1981) Bayesian system identification. In: Eykhoff P. (eds) Trends and Progress in System Identification. Pergamon Press, Oxford, pp 239–304
Poole A., Raftery A. (2000) Inference for deterministic simulation models: The Bayesian melding approach. Journal of the American Statistical Association 95: 1244–1255
Schervish M. J. (1995) Theory of statistics. Springer, New York
Smith J. Q. (1979) A generalisation of the Bayesian steady forecasting model. Journal of the Royal Statistical Society B 41: 375–387
Smith J.Q. (2007) Local robustness of Bayesian parametric inference and observed likelihoods CRiSM. Research Report 07-09. University of Warwick, UK
Tong Y.L. (1980) Probability inequalities in multivariate distributions. Academic Press, New York
Wasserman L. (1992) Invariance properties of density ratio priors. The Annals of Statistics 20: 2177–2182
West M., Harrison P.J. (1997) Bayesian forecasting and dynamic models. In Springer Series in Statistics. Springer, New York
Author information
Authors and Affiliations
Corresponding author
About this article
Cite this article
Smith, J.Q., Rigat, F. Isoseparation and robustness in parametric Bayesian inference. Ann Inst Stat Math 64, 495–519 (2012). https://doi.org/10.1007/s10463-011-0334-9
Received:
Revised:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10463-011-0334-9