Predictive Inference Based on Markov Chain Monte Carlo Output
From MaRDI portal
Publication:6088268
DOI10.1111/INSR.12405arXiv1608.06802OpenAlexW3088024399MaRDI QIDQ6088268FDOQ6088268
Thordis L. Thorarinsdottir, Fabian Krüger, Tilmann Gneiting, Sebastian Lerch
Publication date: 13 December 2023
Published in: International Statistical Review (Search for Journal in Brave)
Abstract: In Bayesian inference, predictive distributions are typically in the form of samples generated via Markov chain Monte Carlo (MCMC) or related algorithms. In this paper, we conduct a systematic analysis of how to make and evaluate probabilistic forecasts from such simulation output. Based on proper scoring rules, we develop a notion of consistency that allows to assess the adequacy of methods for estimating the stationary distribution underlying the simulation output. We then provide asymptotic results that account for the salient features of Bayesian posterior simulators, and derive conditions under which choices from the literature satisfy our notion of consistency. Importantly, these conditions depend on the scoring rule being used, such that the choices of approximation method and scoring rule are intertwined. While the logarithmic rule requires fairly stringent conditions, the continuous ranked probability score (CRPS) yields consistent approximations under minimal assumptions. These results are illustrated in a simulation study and an economic data example. Overall, mixture-of-parameters approximations which exploit the parametric structure of Bayesian models perform particularly well. Under the CRPS, the empirical distribution function is a simple and appealing alternative option.
Full work available at URL: https://arxiv.org/abs/1608.06802
Cites Work
- Title not available (Why is that?)
- Title not available (Why is that?)
- Title not available (Why is that?)
- Title not available (Why is that?)
- Title not available (Why is that?)
- Title not available (Why is that?)
- Coherent dispersion criteria for optimal experimental design
- Understanding predictive information criteria for Bayesian models
- Bayesian data analysis.
- Asymptotic equivalence of Bayes cross validation and widely applicable information criterion in singular learning theory
- Bayesian Measures of Model Complexity and Fit
- Strictly Proper Scoring Rules, Prediction, and Estimation
- Predictive Model Assessment for Count Data
- Markov chains for exploring posterior distributions. (With discussion)
- Comparing Density Forecasts Using Threshold- and Quantile-Weighted Scoring Rules
- The Calculation of Posterior Distributions by Data Augmentation
- A New Approach to the Economic Analysis of Nonstationary Time Series and the Business Cycle
- Bayesianly justifiable and relevant frequency calculations for the applied statistician
- Present Position and Potential Developments: Some Personal Views: Statistical Theory: The Prequential Approach
- Remarks on Some Nonparametric Estimates of a Density Function
- Sampling-Based Approaches to Calculating Marginal Densities
- UNIFORM CONVERGENCE RATES FOR KERNEL ESTIMATION WITH DEPENDENT DATA
- Asymptotic Statistics
- On Choosing and Bounding Probability Metrics
- Bayes Factors
- All of Nonparametric Statistics
- Optimal Transport
- Robust Statistics
- Tests of Conditional Predictive Ability
- Using Proper Divergence Functions to Evaluate Climate Models
- On Kullback-Leibler loss and density estimation
- Nonparametric curve estimation from time series
- Expected information as ecpected utility
- Predictive fit for natural exponential families
- Minorization Conditions and Convergence Rates for Markov Chain Monte Carlo
- Contemporary Bayesian Econometrics and Statistics
- The Deviance Information Criterion: 12 Years on
- Admissible probability measurement procedures
- Nonparameteric estimation in mixing sequences of random variables
- Strong convergence of sums of \(\alpha \)-mixing random variables with applications to density estimation
- Stochastic volatility models as hidden Markov models and statistical applications
- Scoring Rules for Continuous Probability Distributions
- On correcting for variance inflation in kernel density estimation
- Rejoinder on: Assessing probabilistic forecasts of multivariate quantities, with an application to ensemble predictions of surface winds
- Distribution estimation consistent in total variation and in two types of information divergence
- A kernel estimator for discrete distributions
- Density estimation in the \(L^ \infty\) norm for dependent data with applications to the Gibbs sampler
- Normal scale mixtures and dual probability densities
- On bandwidth choice for density estimation with dependent data
- Use and communication of probabilistic forecasts
- Density Estimation for the Metropolis–Hastings Algorithm
- Space and circular time log Gaussian Cox processes with application to crime event data
- Multivariate Type G Matérn Stochastic Partial Differential Equation Random Fields
- Introduction to Bayesian Econometrics
Cited In (6)
- Dynamic pricing using flexible heterogeneous sales response models
- Empirical Bayes inference for the block maxima method
- Bayesian cross-validation by parallel Markov chain Monte Carlo
- Title not available (Why is that?)
- Prediction intervals for economic fixed-event forecasts
- Validation of point process predictions with proper scoring rules
Recommendations
- Multivariate output analysis for Markov chain Monte Carlo 👍 👎
- Markov Chain Monte Carlo 👍 👎
- MCMC methods to approximate conditional predictive distributions 👍 👎
- Predictive characterization of mixtures of Markov chains 👍 👎
- Title not available (Why is that?) 👍 👎
- Title not available (Why is that?) 👍 👎
- Title not available (Why is that?) 👍 👎
- Title not available (Why is that?) 👍 👎
This page was built for publication: Predictive Inference Based on Markov Chain Monte Carlo Output
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6088268)
