The posterior predictive distribution is the distribution of future observations, conditioned on the information available from existing observations. It is the main Bayesian tool for treating predictive problems in statistics. We define the posterior predictive distribution and illustrate its main features in Bayesian parametric inference. We also focus on predictive model checking and selection, which are procedures for checking model adequacy and for selecting a model, when the analysis is based on a posterior predictive approach.
Barbieri, M.M. (2015). Posterior Predictive Distribution. In Wiley StatsRef: Statistics Reference Online (pp. 1-6). Wiley [10.1002/9781118445112.stat07839].
Posterior Predictive Distribution
BARBIERI, Maria Maddalena
2015-01-01
Abstract
The posterior predictive distribution is the distribution of future observations, conditioned on the information available from existing observations. It is the main Bayesian tool for treating predictive problems in statistics. We define the posterior predictive distribution and illustrate its main features in Bayesian parametric inference. We also focus on predictive model checking and selection, which are procedures for checking model adequacy and for selecting a model, when the analysis is based on a posterior predictive approach.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
