Bayesian Analysis

Comment on Article by Müller and Mitra

Peter D. Hoff

Full-text: Open access


Due to their great flexibility, nonparametric Bayes methods have proven to be a valuable tool for discovering complicated patterns in data. The term “nonparametric Bayes” suggests that these methods inherit model-free operating characteristics of classical nonparametric methods, as well as coherent uncertainty assessments provided by Bayesian procedures. However, as the authors say in the conclusion to their article, nonparametric Bayesian methods may be more aptly described as “massively parametric.” Furthermore, I argue that many of the default nonparametric Bayes procedures are only Bayesian in the weakest sense of the term, and cannot be assumed to provide honest assessments of uncertainty merely because they carry the Bayesian label. However useful such procedures may be, we should be cautious about advertising default nonparametric Bayes procedures as either being “assumption free” or providing descriptions of our uncertainty. If we want our nonparametric Bayes procedures to have a Bayesian interpretation, we should modify default NP Bayes methods to accommodate real prior information, or at the very least, carefully evaluate the effects of hyperparameters on posterior quantities of interest.

Article information

Bayesian Anal., Volume 8, Number 2 (2013), 311-318.

First available in Project Euclid: 24 May 2013

Permanent link to this document

Digital Object Identifier

Mathematical Reviews number (MathSciNet)

Zentralblatt MATH identifier

marginal likelihood model misspecification prior specification sandwich estimation


Hoff, Peter D. Comment on Article by Müller and Mitra. Bayesian Anal. 8 (2013), no. 2, 311--318. doi:10.1214/13-BA811B.

Export citation


  • Bush, C., Lee, J., and MacEachern, S. (2010). “Minimially informative prior distributions for non-parametric Bayesian analysis.” Journal of the Royal Statistical Society B, 72(2): 253–268.
  • Efron, B. (1993). “Bayes and likelihood calculations from confidence intervals.” Biometrika, 80(1): 3–26.
  • Ghosal, S. (2001). “Convergence rates for density estimation with Bernstein polynomials.” Annals of Statististics, 29(5): 1264–1280.
  • Greco, L., Racugno, W., and Ventura, L. (2008). “Robust likelihood functions in Bayesian inference.” Journal of Statistical Planning and Inference, 138(5): 1258–1270.
  • Hoff, P. and Wakefield, J. (2012). “Bayesian sandwich posteriors for pseudo-true parameters.” Technical Report 605, Department of Statistics, University of Washington.
  • Hoff, P. D. (2007). “Extending the rank likelihood for semiparametric copula estimation.” Annals of Applied Statistics, 1(1): 265–283.
  • Huber, P. (1967). “The behavior of maximum likelihood estimators under non-standard conditions.” In LeCam, L. and Neyman, J. (eds.), Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability, 221–233. University of California Press.
  • Kessler, D., Hoff, P., and Dunson, D. (2012). “Marginally Specified Priors for Nonparametric Bayesian Estimation.” Technical Report 596, Department of Statistics, University of Washington.
  • Lazar, N. A. (2003). “Bayesian empirical likelihood.” Biometrika, 90(2): 319–326.
  • Lijoi, A. and Regazzini, E. (2004). “Means of a Dirichlet Process and Multiple Hypergeometric Functions.” The Annals of Probability, 32(2): 1469–1495.
  • Müller, P. and Mitra, R. (2013). “Bayesian Nonparametric Inference - Why and How.” Bayesian Analysis. To appear.
  • Müller, U. (2012). “Risk of Bayesian inference in misspecified models, and the sandwich covariance matrix.” Unpublished Manuscript, Princeton University.
  • Severini, T. A. (1991). “On the relationship between Bayesian and non-Bayesian interval estimates.” Journal of the Royal Statistical Society, B, 53(3): 611–618.
  • Szpiro, A. A., Rice, K. M., and Lumley, T. (2010). “Model-robust regression and a Bayesian “sandwich” estimator.” Annals of Applied Statistics, 4(4): 2099–2113.
  • Yamato, H. (1984). “Characteristic functions of means of distributions chosen from a Dirichlet process.” Annals of Probability, 12(1): 262–267.

See also

  • Related item: Peter Müller, Riten Mitra. Bayesian Nonparametric Inference – Why and How. Bayesian Anal., Vol. 8, Iss. 2 (2013) 269–302.