Bayesian Inference with Specified Prior Marginals
We show how to find bounds on posterior expectations of arbitrary functions of the parameters when the prior marginals are specified but when the complete joint prior is unspecified. We also give a theorem that is useful for finding posterior bounds in a wide range of Bayesian robustness problems. We apply these techniques to two examples. The first example involves a recent clinical trial for extracorporeal membrane oxygenation (ECMO). The second example is the product of means problem, which has been studied in the Bayesian context by \textit{J. O. Berger} and \textit{J. M. Bernardo} [ibid. 84, No. 405, 200--207 (1989; Zbl 0682.62018)]. Here the goal is to find the posterior expectation of \(\alpha\beta\), where \(\alpha\) and \(\beta\) are the means of conditionally independent random variables \(X\) and \(Y\). Berger and Bernardo recommended a joint prior \(\pi_ 0\) proportional to \((\alpha^ 2+\beta^ 2)^{1/2}\). We find that among all priors with the same marginals as \(\pi_ 0\), the posterior expectation of \(\alpha\beta\) can be made arbitrarily large or arbitrarily close to \(0\). Furthermore, the parameterization is important: with a different parameterization the upper bound is strictly finite.
- Marginally specified priors for non-parametric Bayesian estimation
- Bayesian Inference Under Partial Prior Information
- scientific article; zbMATH DE number 872117
- Numerical Bayesian inference with arbitrary prior
- Bayesian estimation of incomplete data using conditionally specified priors
- Bayesian learning for a class of priors with prescribed marginals
- Approximately Bayesian Inference
- Range of the posterior probability of an interval for priors with unimodality preserving contaminations
- A likelihood based robust Bayesian summary
- Bayesian analysis under \(\varepsilon\)-contaminated priors: A trade-off between robustness and precision
- Linearization of Bayesian robustness problems
- Bayesian robustness in bidimensional models: Prior independence. (With discussion)
- An overview of robust Bayesian analysis. (With discussion)
- Computing posterior upper expectations
- Classes of bidimensional priors specified on a collection of sets: Bayesian robustness
- Robust Bayesian diagnostics
- Inferences for hierarchical models with partial prior information
- Bayesian learning for a class of priors with prescribed marginals
- Credal networks under epistemic irrelevance
- Adjusted likelihoods for synthesizing empirical evidence from studies that differ in quality and design: effects of environmental tobacco smoke
- Prior Density Selection as a Particular Case of Bayesian Model Selection: A Predictive Approach
- Block unimodality for multivariate Bayesian robustness
- Bayesian Robustness of the Posterior Predictivep-Value
- On the independence between risk profiles in the compound collective risk actuarial model
- Estimating a Product of Means: Bayesian Analysis with Reference Priors
- Bayesian robustness of credible regions in the presence of nuisance parameters
- Sensitivity to prior independence via farlie-gumbel-morgenstern model
This page was built for publication: Bayesian Inference with Specified Prior Marginals
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4031117)
