Model-Robust Bayesian Regression and the Sandwich Estimator
Download the Paper Forward to a colleague
Abstract:
In applied regression problems there is often sufficient data for accurate estimation, but standard parametric models do not accurately describe the source of the data, so associated uncertainty estimates are not reliable. We describe a simple Bayesian approach to inference in linear regression that recovers least-squares point estimates while providing correct uncertainty bounds by explicitly recognizing that standard modeling assumptions need not be valid. Our model-robust development parallels frequentist estimating equations and leads to intervals with the same robustness properties as the ’sandwich’ estimator.
Subject Area:
Statistical Theory and Methods
Suggested Citation:
Adam A. Szpiro, Kenneth M. Rice, and Thomas Lumley, "Model-Robust Bayesian Regression and the Sandwich Estimator" (December 26, 2007). UW Biostatistics Working Paper Series. Working Paper 320.
http://www.bepress.com/uwbiostat/paper320