ABSTRACT
This chapter introduces the Bayesian approach, upon which the majority of the methodology is based, while making comparisons to the likelihood approach. In Bayesian statistics, probability describes all types of uncertainty, both through unpredictability and through imperfect knowledge. A posterior distribution captures our beliefs about a particular quantity and will be the basis of interpretation and possible further inference. Posterior quantiles are invariant to transformations. The set of unknown parameters in a model may include some that are not of primary interest and could be considered nuisance parameters. Conjugate priors are more difficult to work with for more than one parameter, as they are not generally independent.
