ABSTRACT
Bayesian inference is about conditioning models to the available data and obtaining posterior distributions. We can do this using pen and paper, computers, or other devices. Additionally we can include, as part of the inference process, the computation of other quantities like the prior and posterior predictive distributions. Bayesian modeling is wider than inference and other equally important tasks are needed for a successful Bayesian data analysis. In this chapter we will discuss some of these tasks and related tools, such as Diagnosing the quality of the inference results obtained using numerical methods. Model criticism, including evaluations of both model assumptions and model predictions. Comparison of models, including model selection or model averaging. Preparation of the results for a particular audience.
