On recursive Bayesian predictive distributions
From MaRDI portal
Publication:4559693
Abstract: A Bayesian framework is attractive in the context of prediction, but a fast recursive update of the predictive distribution has apparently been out of reach, in part because Monte Carlo methods are generally used to compute the predictive. This paper shows that online Bayesian prediction is possible by characterizing the Bayesian predictive update in terms of a bivariate copula, making it unnecessary to pass through the posterior to update the predictive. In standard models, the Bayesian predictive update corresponds to familiar choices of copula but, in nonparametric problems, the appropriate copula may not have a closed-form expression. In such cases, our new perspective suggests a fast recursive approximation to the predictive density, in the spirit of Newton's predictive recursion algorithm, but without requiring evaluation of normalizing constants. Consistency of the new algorithm is shown, and numerical examples demonstrate its quality performance in finite-samples compared to fully Bayesian and kernel methods.
Recommendations
Cites work
- scientific article; zbMATH DE number 469135 (Why is no real title available?)
- scientific article; zbMATH DE number 1972910 (Why is no real title available?)
- scientific article; zbMATH DE number 774881 (Why is no real title available?)
- A Bayesian analysis of some nonparametric problems
- A model for association in bivariate life tables and its application in epidemiological studies of familial tendency in chronic disease incidence
- A recursive algorithm for nonparametric analysis with missing data
- A survey of sequential Monte Carlo methods for economics and finance
- An introduction to copulas. Properties and applications
- Asymptotic properties of predictive recursion: robustness and rate of convergence
- Autoregressive mixture models for dynamic spatial Poisson processes: application to tracking intensity of violent crime
- Bayesian Density Estimation and Inference Using Mixtures
- Consistency of a recursive estimate of mixing distributions
- Efficient on-line nonparametric kernel density estimation
- Nonparametric density estimation of streaming data using orthogonal series
- On a class of Bayesian nonparametric estimates: I. Density estimates
- On a nonparametric recursive estimator of the mixing distribution
- Particle learning for general mixtures
- Semiparametric inference in mixture models with predictive recursion marginal likelihood
- Sequential Monte Carlo for Bayesian sequentially designed experiments for discrete data
- Stochastic approximation and Newton's estimate of a mixing distribution
- Symmetric Measures on Cartesian Products
Cited in
(14)- Valid inferential models for prediction in supervised learning problems
- scientific article; zbMATH DE number 3913501 (Why is no real title available?)
- A survey of nonparametric mixing density estimation via the predictive recursion algorithm
- Copula particle filters
- Asymptotics of certain conditionally identically distributed sequences
- Recursive equations for the predictive distributions of some determinantal processes
- Sample Size Determination for Credibility Estimation
- Revisiting consistency of a recursive estimator of mixing distributions
- Bayesian Predictive Inference Without a Prior
- Online learning for the Dirichlet process mixture model via weakly conjugate approximation
- A class of models for Bayesian predictive inference
- Approximate Bayesian recursive estimation
- Kernel based Dirichlet sequences
- scientific article; zbMATH DE number 927328 (Why is no real title available?)
This page was built for publication: On recursive Bayesian predictive distributions
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4559693)
