Year: 2025
Type: article
Abstract: This paper reviews the growing field of Bayesian prediction. Bayesian point and interval prediction are defined and situated in statistical prediction more generally. Then, four general approaches to Bayesian prediction are described and we turn to predictor selection. This can be done predictively or nonpredictively; predictors can be based on single models or multiple models. We call these unitary predictors and model average predictors, respectively. Predictors can also be explanatory meaning they are based on physical modeling or algorithmic meaning they are regarded simply as input-output relations. Then we turn to aspects of prediction in the context of large observational data sets, the most recent topic to emerge in prediction. We conclude with a summary and statement of several open problems.
Source: Statistical Science
DOI: 10.1214/24-sts951
Authors: Bertrand Clarke, Yuling Yao
Language: English
Subfield: Statistics and Probability
Field: Mathematics
Domain: Physical Sciences
Open Access status: bronze
OpenAlex ID: W4407099920