Bayesian inference with misspecified models

By Stephen G. Walker. 2013

103 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/b79731d71b20de366a20321a8dddec0fef9d75d5

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

Citations in the corpus

Models as Approximations I: Consequences Illustrated with Linear Regression (2014, background): “…developments: Szpiro, Rice and Lumley (2010) derive a sandwich estimator from Bayesian assumptions, and a lively discussion of misspecification from a Bayesian perspective involved Walker (2013), De Blasi (2013), Hoff and Wakefield (2013) and O’Hagan (2013), who provide further references.”
Models as Approximations - A Conspiracy of Random Regressors and Model Deviations Against Classical Inference in Regression (2015, background): “…are taking place: Szpiro, Rice and Lumley (2010) derive a sandwich estimator from Bayesian assumptions, and a lively discussion of misspecification from a Bayesian perspective involved Walker (2013), De Blasi (2013), Hoff and Wakefield (2013) and O’Hagan (2013), who provide further references.”
Martingale posterior distributions (2021, background): “The choice of the negative log–likelihood is also interesting as it allows us to target the parameters of a parametric model without the assumption that the model is well–specified (Walker, 2013; Bissiri et al., 2016).”
Inconsistency of Bayesian Inference for Misspecified Linear Models, and a Proposal for Repairing It (2014, background): “Indeed, the strongest actual theorems to this end that we know of, (Kleijn and Van der Vaart, 2006, De Blasi and Walker, 2013, Ramamoorthi et al., 2015), hold, as the authors emphasize, under regularity conditions that are substantially stronger than those needed for consistency when the model is…”