A Bayesian “Sandwich” for Variance Estimation Large-sample Bayesian analogs exist for many frequentist methods, but are less well known for the widely-used “sandwich” or “robust” variance estimators. We review existing approaches to Bayesian analogs of sandwich variance estimators and propose a new analog, as the Bayes rule under a form of balanced loss function, that combines elements of standard parametric inference with fidelity of the data to the model. Our development is general, for essentially any regression setting with independent outcomes. Being the large-sample equivalent of its frequentist counterpart, we show by simulation that Bayesian robust standard error estimates can faithfully quantify the variability of parameter estimators even under model misspecification—thus retaining the major attraction of the original frequentist version. We demonstrate our Bayesian analog of standard error estimates when studying the association between age and systolic blood pressure in NHANES.

An alternative approach (penalizing poor fit instead of summarizing a very flexible model, doi.org/10.1214/24-S...) may make it easier to adapt between large-sample robustness and small-sample stability. Still needs work, though...

Only small enough to fit in a widely-quoted statistician, who the Beeb clearly thinks is outdated

Hi there! Just to confuse things further I'm going to throw in lots of discussion of post-2017 work ;)

I'm talking on meta-analysis next Tuesday (5:30pm UK, 9:30am Seattle, etc) with discussion from Peter Diggle and others - should be fun - please do join us.

Very good! (...I am always satisfied with the best)

Win one of our beta hats! Unstructured chino twill cap. Navy blue with beta symbol in white, off-center. White Stata logo on back. Adjustable strap.

Seems the right place to show off a www.stata.com/giftshop/bet...

Hear from Kenneth Rice on his paper 'A Re-Evaluation of Fixed Effect(s) Meta-Analysis’ at our next journal webinar this month💻

Chaired by David Hoaglin and in discussion with Peter Diggle

📆 Tues 24 June
⏱️ 17:30-18:30 BST

rss.org.uk/training-eve...

Sorry to hear this. I don't have a million but will DM you...