Abstract
We propose a geometric framework to assess global sensitivity in Bayesian nonparametric models for density estimation. We study sensitivity of nonparametric Bayesian models for density estimation, based on Dirichlet-type priors, to perturbations of either the precision parameter or the base probability measure. To quantify the different effects of the perturbations of the parameters and hyperparameters in these models on the posterior, we define three geometrically-motivated global sensitivity measures based on geodesic paths and distances computed under the nonparametric Fisher-Rao Riemannian metric on the space of densities, applied to posterior samples of densities: (1) the Fisher-Rao distance between density averages of posterior samples, (2) the log-ratio of Karcher variances of posterior samples, and (3) the norm of the difference of scaled cumulative eigenvalues of empirical covariance operators obtained from posterior samples. We validate our approach using multiple simulation studies, and consider the problem of sensitivity analysis for Bayesian density estimation models in the context of three real datasets that have previously been studied.
Similar content being viewed by others
References
Bean, A., Xu, X. and MacEachern, S. (2016). Transformations and Bayesian density estimation. Electr. J. Statist. 10, 2, 3355–3373.
Berger, J.O. (1982). The robust Bayesian viewpoint, Technical Report, Purdue University Department of Statistics.
Berger, J.O. (1990). Robust Bayesian analysis: Sensitivity to the prior. Journal of Statistical Planning and Inference 25, 3, 303–328.
Berger, J.O. (1994). An overview of robust Bayesian analysis. Test 3, 1, 5–124.
Bhattacharyya, A. (1943). On a measure of divergence between two statistical populations defined by their probability distribution. Bull. Calcutta Math. Soc. 35, 99–109.
Blackwell, D. and MacQueen, J.B. (1973). Ferguson distributions via Pólya urn schemes. Ann. Statist. 1, 2, 353–355.
Bush, C.A., Lee, J. and MacEachern, S.N. (2010). Minimally informative prior distributions for non-parametric Bayesian analysis. J. R. Statist. Soc. Series B 72, 2, 253–268.
Cencov, N.N. (1982). Statistical decision rules and optimal inference, no. 53. American Mathematical Society.
Dey, D.D., Müller, P. and Sinha, D. (1998). Practical nonparametric and semiparametric Bayesian statistics, vol. 133 Springer.
Dryden, I.L. and Mardia, K.V. (1998). Statistical shape analysis, 4. Wiley, Chichester.
Escobar, M.D. (1994). Estimating normal means with a Dirichlet process prior. J. Am. Stat. Assoc. 89, 425, 268–277.
Escobar, M.D. and West, M. (1995). Bayesian density estimation and inference using mixtures. J. Am. Stat. Assoc. 90, 430, 577–588.
Fearnhead, P. (2004). Particle filters for mixture models with an unknown number of components. Stat. Comput. 14, 1, 11–21.
Ferguson, T.S. (1973). A Bayesian analysis of some nonparametric problems. Ann. Statist. 1, 209–230.
Ferguson, T.S. (1983). Bayesian density estimation by mixtures of normal distributions. In Recent advances in statistics. Elsevier, pp. 287–302.
Gershman, S.J. and Blei, D.M. (2012). A tutorial on Bayesian nonparametric models. J. Math. Psychol. 56, 1, 1–12.
Ghosal, S. and Van Der Vaart, A.W. (2001). Entropies and rates of convergence for maximum likelihood and Bayes estimation for mixtures of normal densities. Ann. Statist. 29, 5, 1233–1263.
Ghosh, J.K. and Ramamoorthi, R.V. (2003). Bayesian nonparametrics. Springer.
Görür, D. and Rasmussen, C.E. (2010). Dirichlet process Gaussian mixture models: Choice of the base distribution. J. Comput. Sci. Technol. 25, 4, 653–664.
Griffin, J.E. (2010). Default priors for density estimation with mixture models. Bayesian Anal. 5, 1, 45–64.
Griffin, J.E. and Steel, M.F. (2004). Semiparametric Bayesian inference for stochastic frontier models. J. Econometr. 123, 1, 121–152.
Gustafson, P. (1996). Local sensitivity of inferences to prior marginals. J. Am. Stat. Assoc. 91, 434, 774–781.
Gustafson, P. (2000). Local robustness in Bayesian analysis. In Robust Bayesian analysis. Springer, pp. 71–88.
Gustafson, P. and Wasserman, L. (1995). Local sensitivity diagnostics for Bayesian inference. Ann. Statist. 23, 6, 2153–2167.
Hjort, N.L., Holmes, C., Müller, P. and Walker, S.G. (2010). Bayesian nonparametrics, vol. 28. Cambridge University Press.
Insua, D.R. and Ruggeri, F. (2012). Robust Bayesian analysis, vol. 152. Springer.
Ishwaran, H. and James, L.F. (2001). Gibbs sampling methods for stick-breaking priors. J. Am. Stat. Assoc. 96, 453, 161–173.
Kass, R.E. and Vos, P.W. (1997). Geometrical foundations of asymptotic inference. Wiley.
Kass, R.E., Tierney, L. and Kadane, J.B. (1989). Approximate methods for assessing influence and sensitivity in Bayesian analysis. Biometrika 76, 4, 663–674.
Kurtek, S. (2017). A geometric approach to pairwise Bayesian alignment of functional data using importance sampling. Electronic Journal of Statistics 11, 1, 502–531.
Kurtek, S. and Bharath, K. (2015). Bayesian sensitivity analysis with Fisher–Rao metric. Biometrika 102, 3, 601–616.
Lang, S. (2012). Fundamentals of differential geometry, vol. 191. Springer Science & Business Media.
Lee, J., MacEachern, S.N., Lu, Y., Mills, G.B. et al. (2014). Local-mass preserving prior distributions for nonparametric Bayesian models. Bayesian Anal. 9, 2, 307–330.
MacEachern, S.N. (1994). Estimating normal means with a conjugate style Dirichlet process prior. Commun. Stat. Simul. Comput. 23, 3, 727–741.
MacEachern, S.N. (1998). Computational methods for mixture of Dirichlet process models. In Practical nonparametric and semiparametric Bayesian statistics. Springer, pp. 23–43.
MacEachern, S.N. (2016). Nonparametric Bayesian methods: A gentle introduction and overview. Communications for Statistical Applications and Methods 23, 6, 445–466.
MacEachern, S.N. and Müller, P. (1998). Estimating mixture of Dirichlet process models. J. Comput. Graph. Stat. 7, 2, 223–238.
MacEachern, S.N., Clyde, M. and Liu, J.S. (1999). Sequential importance sampling for nonparametric Bayes models: The next generation. Can. J. Stat. 27, 2, 251–267.
Millar, R.B. and Stewart, W.S. (2007). Assessment of locally influential observations in Bayesian models. Bayesian Anal. 2, 2, 365–383.
Müller, P. and Quintana, F.A. (2004). Nonparametric Bayesian data analysis. Stat. Sci. 19, 1, 95–110.
Müller, P. and Rodriguez, A. (2013). Nonparametric Bayesian inference. Institute of Mathematical Statistics.
Müller, P., Quintana, F.A., Jara, A. and Hanson, T. (2015). Bayesian nonparametric data analysis. Springer.
Neal, R.M. (2000). Markov chain sampling methods for Dirichlet process mixture models. J. Comput. Graph. Stat. 9, 2, 249–265.
Newton, M.A. and Zhang, Y. (1999). A recursive algorithm for nonparametric analysis with missing data. Biometrika 86, 1, 15–26.
Nieto-Barajas, L.E. and Prünster, I. (2009). A sensitivity analysis for Bayesian nonparametric density estimators. Stat. Sin. 19, 685–705.
Oakley, J.E. and O’Hagan, A. (2004). Probabilistic sensitivity analysis of complex models: A Bayesian approach. J. R. Statist. Soc. Series B 66, 3, 751–769.
Pitman, J. (2002). Combinatorial stochastic processes, Technical Report 621, University of California, Berkeley, Department of Statistics.
Rao, C.R. (1945). Information and accuracy attainable in the estimation of statistical parameters. Bull. Calcutta Math. Soc. 37, 81–91.
Richardson, S. and Green, P.J. (1997). On Bayesian analysis of mixtures with an unknown number of components (with discussion). J. R. Statist. Soc. Series B 59, 4, 731–792.
Roeder, K. and Wasserman, L. (1997). Practical Bayesian density estimation using mixtures of normals. J. Am. Stat. Assoc. 92, 439, 894–902.
Roos, M., Martins, T.G., Held, L. and Rue, H. (2015). Sensitivity analysis for Bayesian hierarchical models. Bayesian Anal. 10, 2, 321–349.
Ruggeri, F. (2008). Bayesian robustness, European working group. Multiple Criteria Decision Aiding 3, 17, 6.
Ruggeri, F. and Sivaganesan, S. (2000). On a global sensitivity measure for Bayesian inference. Sankhyā Series A 62, 110–127.
Sethuraman, J. (1994). A constructive definition of Dirichlet priors. Stat. Sin. 4, 639–650.
Sivaganesan, S. (2000). Global and local robustness approaches: Uses and limitations. In Robust Bayesian analysis. Springer, pp. 89–108.
Walker, S. (2004). New approaches to Bayesian consistency. Ann. Statist. 32, 5, 2028–2043.
Walker, S.G. and Mallick, B.K. (1997). A note on the scale parameter of the Dirichlet process. Can. J. Stat. 25, 4, 473–479.
Walker, S.G., Damien, P., Laud, P.W. and Smith, A.F. (1999). Bayesian nonparametric inference for random distributions and related functions. J. R. Statist. Soc. Series B 61, 3, 485–527.
Yang, L. and Marron, J.S. (1999). Iterated transformation–kernel density estimation. J. Am. Stat. Assoc. 94, 446, 580–589.
Zhu, H., Ibrahim, J.G. and Tang, N. (2011). Bayesian influence analysis: A geometric approach. Biometrika 98, 2, 307–323.
Acknowledgments
The authors would like to thank Karthik Bharath for valuable discussions and suggestions. SK was partially supported by NSF DMS 1613054, NSF CCF 1740761 and NIH R01-CA214955.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Saha, A., Kurtek, S. Geometric Sensitivity Measures for Bayesian Nonparametric Density Estimation Models. Sankhya A 81, 104–143 (2019). https://doi.org/10.1007/s13171-018-0145-7
Received:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s13171-018-0145-7
Keywords and phrases.
- Global sensitivity analysis
- Fisher–Rao metric
- Bayesian nonparametric density estimation
- square-root density
- Dirichlet process
- Dirichlet process Gaussian mixture model.