Adaptive Bayesian credible sets in regression with a Gaussian process prior
From MaRDI portal
Abstract: We investigate two empirical Bayes methods and a hierarchical Bayes method for adapting the scale of a Gaussian process prior in a nonparametric regression model. We show that all methods lead to a posterior contraction rate that adapts to the smoothness of the true regression function. Furthermore, we show that the corresponding credible sets cover the true regression function whenever this function satisfies a certain extrapolation condition. This condition depends on the specific method, but is implied by a condition of self-similarity. The latter condition is shown to be satisfied with probability one under the prior distribution.
Recommendations
- Adaptive Bayesian credible bands in regression with a Gaussian process prior
- Frequentist coverage of adaptive nonparametric Bayesian credible sets
- Credible sets in the fixed design model with Brownian motion prior
- Adaptive Bayesian estimation using a Gaussian random field with inverse gamma bandwidth
- Bayesian inference with rescaled Gaussian process priors
Cites work
- scientific article; zbMATH DE number 3856278 (Why is no real title available?)
- A Correspondence Between Bayesian Estimation on Stochastic Processes and Smoothing by Splines
- Adaptive Confidence Bands for Nonparametric Regression Functions
- Adaptive confidence balls
- Adaptive confidence bands
- Adaptive nonparametric confidence sets
- An adaptation theory for nonparametric confidence intervals
- An analysis of Bayesian inference for nonparametric regression
- Bayesian inference with rescaled Gaussian process priors
- Bayesian inverse problems with Gaussian priors
- Confidence bands in density estimation
- Convergence rates of posterior distributions.
- Credible sets in the fixed design model with Brownian motion prior
- Empirical Bayes scaling of Gaussian priors in the white noise model
- Frequentist coverage of adaptive nonparametric Bayesian credible sets
- Honest adaptive confidence bands and self-similar functions
- On adaptive inference and confidence bands
- On nonparametric confidence intervals
- On the Bernstein-von Mises theorem with infinite-dimensional parameters
- Rates of contraction of posterior distributions based on Gaussian process priors
- Weak convergence and empirical processes. With applications to statistics
Cited in
(13)- Comment: ``Bayes, oracle Bayes and empirical Bayes
- Empirical Bayes oracle uncertainty quantification for regression
- Adaptive Gaussian Process Approximation for Bayesian Inference with Expensive Likelihood Functions
- Posterior contraction and credible regions for level sets
- Asymptotic frequentist coverage properties of Bayesian credible sets for sieve priors
- Direct Bayesian linear regression for distribution-valued covariates
- Joint non-parametric estimation of mean and auto-covariances for Gaussian processes
- Can we trust Bayesian uncertainty quantification from Gaussian process priors with squared exponential covariance kernel?
- Locally adaptive Bayes nonparametric regression via nested Gaussian processes
- A fast and calibrated computer model emulator: an empirical Bayes approach
- Rejoinder: ``Probabilistic integration: a role in statistical computation?
- Adaptation bounds for confidence bands under self-similarity
- Adaptive Bayesian credible bands in regression with a Gaussian process prior
This page was built for publication: Adaptive Bayesian credible sets in regression with a Gaussian process prior
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q895012)
