×

zbMATH — the first resource for mathematics

Sensitivity analysis for Bayesian hierarchical models. (English) Zbl 1335.62059
Summary: Prior sensitivity examination plays an important role in applied Bayesian analyses. This is especially true for Bayesian hierarchical models, where interpretability of the parameters within deeper layers in the hierarchy becomes challenging. In addition, lack of information together with identifiability issues may imply that the prior distributions for such models have an undesired influence on the posterior inference. Despite its importance, informal approaches to prior sensitivity analysis are currently used. They require repetitive re-fits of the model with ad-hoc modified base prior parameter values. Other formal approaches to prior sensitivity analysis suffer from a lack of popularity in practice, mainly due to their high computational cost and absence of software implementation. We propose a novel formal approach to prior sensitivity analysis, which is fast and accurate. It quantifies sensitivity without the need for a model re-fit. Through a series of examples we show how our approach can be used to detect high prior sensitivities of some parameters as well as identifiability issues in possibly over-parametrized Bayesian hierarchical models.

MSC:
62F15 Bayesian inference
62F35 Robustness and adaptive procedures (parametric inference)
Software:
BayesDA; LISP-STAT
PDF BibTeX XML Cite
Full Text: DOI Euclid
References:
[1] Amari, S. (1990). Differential-Geometrical Methods in Statistics. 2nd Edition. Lecture Notes in Statistics , volume 28. Springer-Verlag. · Zbl 0701.62008
[2] Amari, S. and Nagaoka, H. (2000). Methods of Information Geometry . Oxford University Press. · Zbl 0960.62005
[3] Berger, J. O., Ríos Insua, D., and Ruggeri, F. (2000). “Bayesian Robustness.” In Ríos Insua, D. and Ruggeri, F. (eds.), Robust Bayesian Analysis , 1-32. Springer-Verlag. · Zbl 1281.62076
[4] Besag, J., York, J., and Mollié, A. (1991). “Bayesian image restoration, with two applications in spatial statistics.” Annals of the Institute of Statistical Mathematics , 43(1): 1-59. · Zbl 0760.62029
[5] Bhattacharyya, A. (1943). “On a measure of divergence between two statistical populations defined by their probability distributions.” Bulletin of the Calcutta Mathematical Society , 35: 99-109. · Zbl 0063.00364
[6] Box, G. (1980). “Sampling and Bayes’ inference in scientific modelling and robustness.” Journal of the Royal Statistical Society, Series A , 143(4): 383-430. · Zbl 0471.62036
[7] Breslow, N. E. and Clayton, D. G. (1993). “Approximate inference in generalized linear mixed models.” Journal of the American Statistical Association , 88(421): 9-25. · Zbl 0775.62195
[8] Cacuci, D. G., Ionescu-Bujor, M., and Navon, I. M. (2005). Sensitivity and Uncertainty Analysis. Volume II, Applications to Large-Scale Systems . Chapman & Hall. · Zbl 1083.60002
[9] Carlin, B. and Louis, T. (1998). Bayes and Empirical Bayes Methods for Data Analysis . Chapman & Hall/CRC. · Zbl 0871.62012
[10] Clarke, B. and Gustafson, P. (1998). “On the overall sensitivity of the posterior distribution to its inputs.” Journal of Statistical Planning and Inference , 71(1-2): 137-150. · Zbl 0938.62028
[11] Cook, R. (1986). “Assessment of local influence.” Journal of the Royal Statistical Society, Series B , 48(2): 133-169. · Zbl 0608.62041
[12] Dawid, A. (1977). “Further comments on some comments on a paper by Bradley Efron.” The Annals of Statistics , 5(6): 1249. · Zbl 0373.62004
[13] - (1979). “Conditional independence in statistical theory.” Journal of the Royal Statistical Society. Series B (Methodological) , 41(1): 1-31. · Zbl 0408.62004
[14] Dey, D. and Birmiwal, L. (1994). “Robust Bayesian analysis using divergence measures.” Statistics & Probability Letters , 20(4): 287-294. · Zbl 0799.62003
[15] Eberly, L. and Carlin, B. (2000). “Identifiability and convergence issues for Markov chain Monte Carlo fitting of spatial models.” Statistics in Medicine , 19(17-18): 2279-2294.
[16] Evans, M. and Moshonov, H. (2006). “Checking for prior-data conflict.” Bayesian Analysis , 4(1): 893-914. · Zbl 1331.62030
[17] Fong, Y., Rue, H., and Wakefield, J. (2010). “Bayesian inference for generalized linear mixed models.” Biostatistics , 11(3): 397-412.
[18] Geisser, S. (1992). “Bayesian perturbation diagnostics and robustness.” In Goel, P. and Iyengar, N. (eds.), Bayesian Analysis in Statistics and Econometrics , 289-301. Springer-Verlag. · Zbl 0825.62414
[19] - (1993). Predictive Inference: An Introduction . Chapman & Hall, Inc. · Zbl 0824.62001
[20] Gelfand, A. and Sahu, S. (1999). “Identifiability, improper priors, and Gibbs sampling for Generalized Linear Models.” Journal of the American Statistical Association , 94(445): 247-253. · Zbl 1072.62611
[21] Gelfand, A. and Smith, A. (1990). “Sampling-based approaches to calculating marginal densities.” Journal of the American Statistical Association , 85(410): 398-409. · Zbl 0702.62020
[22] Gelman, A., Carlin, J., Stern, H., and Rubin, D. (2004). Bayesian Data Analysis. 2nd Edition . Chapman & Hall/CRC. · Zbl 1039.62018
[23] Gilks, W., Richardson, S., and Spiegelhalter, D. (1996). Markov Chain Monte Carlo . Chapman & Hall. · Zbl 0845.60072
[24] Goutis, C. and Robert, C. (1998). “Model choice in generalised linear models: A Bayesian approach via Kullback-Leibler projections.” Biometrika , 85(1): 29-37. · Zbl 0903.62061
[25] Gustafson, P. (1996). “Local sensitivity of inferences to prior marginals.” Journal of the American Statistical Association , 91(434): 774-781. · Zbl 0869.62022
[26] - (2000). “Local robustness in Bayesian analysis.” In Ríos Insua, D. and Ruggeri, F. (eds.), Robust Bayesian Analysis , 71-88. Springer-Verlag. · Zbl 1281.62068
[27] Gustafson, P. and Wasserman, L. (1995). “Local sensitivity diagnostics for Bayesian inference.” The Annals of Statistics , 23(6): 2153-2167. · Zbl 0854.62024
[28] Ibrahim, J., Zhu, H., and Tang, N. (2011). “Bayesian local influence for survival models.” Lifetime Data Analysis , 17(1): 43-70. · Zbl 1322.62103
[29] Jeffreys, H. (1961). Theory of Probability . Oxford University Press. · Zbl 0116.34904
[30] Kadane, J. (1992). “Comments to: “Bayesian perturbation diagnostics and robustness” by S. Geisser.” In Goel, P. and Iyengar, N. (eds.), Bayesian Analysis in Statistics and Econometrics , 298-300. Springer-Verlag.
[31] Kass, R., Tierney, L., and Kadane, J. (1989). “Approximate methods for assessing influence and sensitivity in Bayesian analysis.” Biometrika , 76(4): 663-674. · Zbl 0678.62037
[32] Lavine, M. (1992). “Local predictive influence in Bayesian linear models with conjugate priors.” Communications in Statistics - Simulation and Computation , 21(1): 269-283. · Zbl 0850.62286
[33] Le Cam, L. (1986). Asymptotic Methods in Statistical Decision Theory . Springer-Verlag. · Zbl 0605.62002
[34] Lesaffre, E. and Lawson, A. (2012). Bayesian Biostatistics . John Wiley & Sons. · Zbl 1282.62057
[35] Lindgren, F., Rue, H., and Lindström, J. (2011). “An explicit link between Gaussian fields and Gaussian Markov random fields: the stochastic differential equation approach.” Journal of the Royal Statistical Society, Series B. , 73(4): 423-498. · Zbl 1274.62360
[36] Martins, T., Simpson, D., Lindgren, F., and Rue, H. (2013). “Bayesian computing with INLA: new features.” Computational Statistics & Data Analysis , 67: 68-83.
[37] McCulloch, R. (1989). “Local model influence.” Journal of the American Statistical Association , 84(406): 473-478.
[38] Millar, R. and Stewart, W. (2007). “Assessment of locally influential observations in Bayesian models.” Bayesian Analysis , 2(2): 365-384. · Zbl 1331.62156
[39] Müller, U. (2012). “Measuring prior sensitivity and prior informativeness in large Bayesian models.” Journal of Monetary Economics , 59(6): 581-597.
[40] Narasimhan, B. (2005). “Lisp-Stat to Java to R.” Journal of Statistical Software , 13(4): 1-10.
[41] Oakley, J. and O’Hagan, A. (2004). “Probabilistic sensitivity analysis of complex models: a Bayesian approach.” Journal of the Royal Statistical Society, Series B , 66(3): 751-769. · Zbl 1046.62027
[42] Pérez, C., Martín, J., and Rufo, M. (2006). “MCMC-based local parametric sensitivity estimation.” Computational Statistics & Data Analysis , 51(2): 823-835. · Zbl 1157.62364
[43] Pettit, L. (1990). “The conditional predictive ordinate for the normal distribution.” Journal of the Royal Statistical Society, Series B , 52(1): 175-184. · Zbl 0699.62031
[44] Plummer, M. (2001). “Local sensitivity in Bayesian graphical models.” URL
[45] Rao, C. (1945). “Information and the accuracy attainable in the estimation of statistical parameters.” Bulletin of the Calcutta Mathematical Society , 37: 81-91. · Zbl 0063.06420
[46] Ríos Insua, D., Ruggeri, F., and Martín, J. (2000). “Bayesian Sensitivity Analysis.” In Saltelli, A., Chan, K., and Scott, E. M. (eds.), Sensitivity Analysis , 225-244. John Wiley & Sons.
[47] Robert, C. (1996). “Intrinsic losses.” Theory and Decision , 40(2): 191-214. · Zbl 0848.90010
[48] Roos, M. and Held, L. (2011). “Sensitivity analysis in Bayesian generalized linear mixed models for binary data.” Bayesian Analysis , 6(2): 259-278. · Zbl 1330.62150
[49] Rue, H., Martino, S., and Chopin, N. (2009). “Approximate Bayesian inference for latent Gaussian models by using integrated nested Laplace approximations.” Journal of the Royal Statistical Society, Series B. , 71(2): 319-392. · Zbl 1248.62156
[50] Ruggeri, F. (2008). “Bayesian Robustness.” Forum: Robustness Analysis, In: European Working Group “Multiple Criteria Decision Aiding” , 3(17).
[51] Saltelli, A., Ratto, M., Andres, T., Campolongo, F., Cariboni, J., Gatelli, D., Saisana, M., and Tarantola, S. (2008). Global Sensitivity Analysis. The Primer . John Wiley & Sons. · Zbl 1161.00304
[52] Saltelli, A., Tarantola, S., Campolongo, F., and Ratto, M. (2004). Sensitivity Analysis in Practice. A Guide to Assessing Scientific Models . John Wiley & Sons. · Zbl 1049.62112
[53] Sivaganesan, S. (2000). “Global and local robustness approaches: uses and limitations.” In Ríos Insua, D. and Ruggeri, F. (eds.), Robust Bayesian Analysis , 89-108. Springer-Verlag. · Zbl 1281.62083
[54] Tierney, L. and Kadane, J. (1986). “Accurate approximations for posterior moments and marginal densities.” Journal of the American Statistical Association , 81(393): 82-86. · Zbl 0587.62067
[55] Van der Linde, A. (2007). “Local influence on posterior distributions under multiplicative modes of perturbation.” Bayesian Analysis , 2(2): 319-332. · Zbl 1331.62171
[56] Weiss, R. (1996). “An approach to Bayesian sensitivity analysis.” Journal of the Royal Statistical Society, Series B. , 58(4): 739-750. · Zbl 0860.62031
[57] Weiss, R. and Cook, R. (1992). “A graphical case statistic for assessing posterior influence.” Biometrika , 79(1): 51-55. · Zbl 0850.62283
[58] Zhu, H., Ibrahim, J., Lee, S., and Zhang, H. (2007). “Perturbation selection and influence measures in local influence analysis.” The Annals of Statistics , 35(6): 2565-2588. · Zbl 1129.62068
[59] Zhu, H., Ibrahim, J., and Tang, N. (2011). “Bayesian influence analysis: a geometric approach.” Biometrika , 98(2): 307-323. · Zbl 1215.62029
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.