Dynamics of Bayesian updating with dependent data and misspecified models
From MaRDI portal
Publication:1952015
Bayesian nonparametricslarge deviationsreplicator dynamicsposterior convergenceBayesian consistencysofic systemsEgorov's theoremasymptotic equipartition
Asymptotic properties of nonparametric inference (62G20) Markov processes: estimation; hidden Markov models (62M05) Large deviations (60F10) Non-Markovian processes: estimation (62M09) Problems related to evolution (92D15) Bayesian problems; characterization of Bayes procedures (62C10) Measures of information, entropy (94A17)
Abstract: Much is now known about the consistency of Bayesian updating on infinite-dimensional parameter spaces with independent or Markovian data. Necessary conditions for consistency include the prior putting enough weight on the correct neighborhoods of the data-generating distribution; various sufficient conditions further restrict the prior in ways analogous to capacity control in frequentist nonparametrics. The asymptotics of Bayesian updating with mis-specified models or priors, or non-Markovian data, are far less well explored. Here I establish sufficient conditions for posterior convergence when all hypotheses are wrong, and the data have complex dependencies. The main dynamical assumption is the asymptotic equipartition (Shannon-McMillan-Breiman) property of information theory. This, along with Egorov's Theorem on uniform convergence, lets me build a sieve-like structure for the prior. The main statistical assumption, also a form of capacity control, concerns the compatibility of the prior and the data-generating process, controlling the fluctuations in the log-likelihood when averaged over the sieve-like sets. In addition to posterior convergence, I derive a kind of large deviations principle for the posterior measure, extending in some cases to rates of convergence, and discuss the advantages of predicting using a combination of models known to be wrong. An appendix sketches connections between these results and the replicator dynamics of evolutionary theory.
Recommendations
- Bayesian updating revisited
- Bayesian model updating with consideration of modeling error
- scientific article; zbMATH DE number 3868426
- scientific article; zbMATH DE number 3848392
- Independence assumptions and Bayesian updating
- scientific article; zbMATH DE number 6142324
- Independence and Bayesian updating methods
- Bayesian inference with misspecified models
- Bayesian updating with confounded signals
- scientific article; zbMATH DE number 1222672
Cites work
- scientific article; zbMATH DE number 1713116 (Why is no real title available?)
- scientific article; zbMATH DE number 5770656 (Why is no real title available?)
- scientific article; zbMATH DE number 48436 (Why is no real title available?)
- scientific article; zbMATH DE number 192937 (Why is no real title available?)
- scientific article; zbMATH DE number 1322222 (Why is no real title available?)
- scientific article; zbMATH DE number 918233 (Why is no real title available?)
- scientific article; zbMATH DE number 2206050 (Why is no real title available?)
- scientific article; zbMATH DE number 3063387 (Why is no real title available?)
- A predictive view of continuous time processes
- A sandwich proof of the Shannon-McMillan-Breiman theorem
- An Introduction to Symbolic Dynamics and Coding
- Bayesian Estimation of the Spectral Density of a Time Series
- Bayesian consistency for stationary models
- Bayesian nonparametrics
- Complexity
- Computational mechanics: pattern and prediction, structure and simplicity.
- Consistency a Posteriori
- Convergence properties of sequential Bayesian \(D\)-optimal designs
- Convergence rates of posterior distributions for non iid observations
- Convergence rates of posterior distributions.
- Correction to: Entropy and the consistent estimation of joint distributions
- Evolutionary Games and Population Dynamics
- Finitary measures for subshifts of finite type and sofic systems
- From \(\varepsilon\)-entropy to KL-entropy: analysis of minimum information complexity density estima\-tion
- From finite to infinite range order via annealing: the causal architecture of deformation faulting in annealed close-packed crystals
- Hidden Markov Models and Dynamical Systems
- How sampling reveals a process
- Large deviations and the Bayesian estimation of higher-order Markov transition functions
- Learning through reinforcement and replicator dynamics
- Likelihood Ratio Tests for Model Selection and Non-Nested Hypotheses
- Limiting Behavior of Posterior Distributions when the Model is Incorrect
- Merging of Opinions with Increasing Information
- Misspecification in infinite-dimensional Bayesian statistics
- Model selection tests for nonlinear dynamic models
- Moderate Deviations for Bayes Posteriors
- New approaches to Bayesian consistency
- Nonparametric maximum likelihood estimation by the method of sieves
- Nonparametric time series prediction through adaptive model selection
- On Bayes procedures
- On Rates of Convergence for Posterior Distributions Under Misspecification
- On the consistency of Bayes estimates
- Prediction, Learning, and Games
- REINFORCEMENT LEARNING IN MARKOVIAN EVOLUTIONARY GAMES
- Rates of convergence of posterior distributions.
- Some PAC-Bayesian theorems
- Some Sufficient Conditions on an Arbitrary Class of Stochastic Processes for the Existence of a Predictor
- Subshifts of finite type and sofic systems
- Sufficient conditions for Bayesian consistency
- Sufficient statistics and extreme points
- Symbolic dynamics. One-sided, two-sided and countable state Markov shifts
- The consistency of posterior distributions in nonparametric problems
- The multiplicative weights update method: a meta-algorithm and applications
- Theory of statistics
Cited in
(27)- Decentralized pricing and the equivalence between Nash and Walrasian equilibrium
- On posterior concentration in misspecified models
- Adaptive Bayesian density estimation with location-scale mixtures
- Optimal point process filtering and estimation of the coalescent process
- General Robust Bayes Pseudo-Posteriors: Exponential Convergence Results with Applications
- On clonal selection
- On classical and Bayesian asymptotics in state space stochastic differential equations
- Non-marginal decisions: a novel Bayesian multiple testing procedure
- Bayesian model comparison with the Hyvärinen score: computation and consistency
- Doob's consistency of a non-Bayesian updating process
- On the brittleness of Bayesian inference
- Experimental design for nonparametric correction of misspecified dynamical models
- Dense orbits of the Bayesian updating group action
- A Short Note on Almost Sure Convergence of Bayes Factors in the General Set-Up
- Gibbs posterior convergence and the thermodynamic formalism
- Darwinian evolution as Brownian motion on the simplex: a geometric perspective on stochastic replicator dynamics
- Examples of inconsistent Bayes procedures based on observations on dynamical systems
- Statistical inference for dynamical systems: a review
- On the importance of the Jacobian determinant in parameter inference for random parameter and random measurement error models
- Bio-inspired feedback-circuit implementation of discrete, free energy optimizing, winner-take-all computations
- On Bayesian learning from Bernoulli observations
- Independence and Bayesian updating methods
- Consistency of maximum likelihood estimation for some dynamical systems
- Runs, panics and bubbles: Diamond-Dybvig and Morris-Shin reconsidered
- scientific article; zbMATH DE number 2202313 (Why is no real title available?)
- Posterior consistency for the spectral density of non‐Gaussian stationary time series
- Asymptotic theory of dependent Bayesian multiple testing procedures under possible model misspecification
This page was built for publication: Dynamics of Bayesian updating with dependent data and misspecified models
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1952015)
