The Annals of Applied Statistics

Model-robust regression and a Bayesian “sandwich” estimator

Adam A. Szpiro, Kenneth M. Rice, and Thomas Lumley

Full-text: Open access


We present a new Bayesian approach to model-robust linear regression that leads to uncertainty estimates with the same robustness properties as the Huber–White sandwich estimator. The sandwich estimator is known to provide asymptotically correct frequentist inference, even when standard modeling assumptions such as linearity and homoscedasticity in the data-generating mechanism are violated. Our derivation provides a compelling Bayesian justification for using this simple and popular tool, and it also clarifies what is being estimated when the data-generating mechanism is not linear. We demonstrate the applicability of our approach using a simulation study and health care cost data from an evaluation of the Washington State Basic Health Plan.

Article information

Ann. Appl. Stat., Volume 4, Number 4 (2010), 2099-2113.

First available in Project Euclid: 4 January 2011

Permanent link to this document

Digital Object Identifier

Mathematical Reviews number (MathSciNet)

Zentralblatt MATH identifier

Bayesian inference estimating equations linear regression robust regression sandwich estimator


Szpiro, Adam A.; Rice, Kenneth M.; Lumley, Thomas. Model-robust regression and a Bayesian “sandwich” estimator. Ann. Appl. Stat. 4 (2010), no. 4, 2099--2113. doi:10.1214/10-AOAS362.

Export citation


  • Crainiceanu, C., Ruppert, D. and Wand, M. P. (2005). Bayesian analysis for penalized spline regression using WinBUGS. J. Statist. Soft. 14 1–24.
  • Davison, A. C. and Hinkley, D. V. (1997). Bootstrap Methods and Their Applications. Univ. Cambridge Press.
  • Diehr, P., Madden, C., Martin, D. P., Patrick, D. L. and Mayers, M. (1993). Who enrolled in a state program for the uninsured: Was there adverse selection? Medical Care 31 1093–1105.
  • Diehr, P., Yanez, D., Ash, A. and Hornbrook, M. (1999). Methods for analyzing health care utilization and costs. Annual Review of Public Health 20 125–144.
  • Freedman, D. A. (2006). On the so-called ‘Huber sandwich estimator’ and ‘robust standard errors.’ Amer. Statist. 60 299–302.
  • Gasparini, M. (1995). Exact multivariate Bayesian bootstrap distributions of moments. Ann. Statist. 23 762–768.
  • Gauss, C. F. (1809). Theoria Motus Corporum Coelestium in Sectionibus Conicis Solem Ambientum. Friedrich Perthes and I.H. Besser, Hamburg.
  • Huber, P. J. (1967). The behavior of maximum likelihood estimates under nonstandard conditions. In Proceedings of the Fifth Berkeley Symposium on Mathematical Statistic and Probability 1 221–233.
  • Jacobi, C. G. J. (1841). De formatione et proprietatibus determinantum. J. Reine Angew. Math. 22 285–318.
  • Legendre, A. M. (1805). Nouvelles Methodes Pour la Determination des Orbites des Cometes. F. Didot, Paris.
  • Liang, K. Y. and Zeger, S. A. (1986). Longitudinal data analysis using generalized linear models. Biometrika 73 13–22.
  • Lumley, T., Diehr, P., Emerson, S. and Chen, L. (2002). The importance of the normality assumption in large public health data sets. Annual Review of Public Health 23 151–169.
  • Lunn, D. J., Thomas, A. and Best, N. (2000). WinBUGS—a Bayesian modelling framework: Concepts, structure, and extensibility. Statist. Comput. 10 325–337.
  • Rice, K., Lumley, T. and Szpiro, A. (2008). Trading bias for precision: Decision theory for intervals and sets. UW Biostatistics Working Paper Series, Working Paper 336.
  • Royall, R. M. (1986). Model robust confidence intervals using maximum likelihood estimators. Int. Statist. Rev. 54 221–226.
  • Rubin, D. B. (1981). The Bayesian bootstrap. Ann. Statist. 9 130–134.
  • Szpiro, A. A., Rice, K. M. and Lumley, T. (2010). Supplement to “Model robust regression and a Bayesian ‘sandwich’ estimator.” DOI: 10.1214/10-AOAS362SUPP.
  • Wand, M. P. and Ormerod, J. T. (2008). On semiparametric regression with O’Sullivan penalised splines. Aust. N.Z. J. Statist. 50 179–198.
  • White, H. (1980). A heteroskedasticity-consistent covariance matrix estimator and a direct test for heteroskedasticity. Econometrica 48 817–838.

Supplemental materials

  • Supplementary material: Proofs of theorems in “Model robust regression and a Bayesian ‘sandwich’ estimator” (Szpiro, Rice, and Lumley). We provide proofs of the theorems stated in the paper “Model robust regression and a Bayesian ‘sandwich’ estimator” by Adam A. Szpiro, Kenneth M. Rice and Thomas Lumley.