The Bayesian Paradigm
- 1.9k Downloads
In the previous chapters, we regarded the population quantities as unknown fixed constants and the observations or records as outcomes of a random process. In Chapter 3 the sampling process and in Chapters 1 and 2 the data-generating process, as described by a model equation or a class of joint distributions, were the sole sources of randomness. This chapter introduces a radically different approach in which the observed quantities (data) are fixed and all unknown quantities are random and described by their joint posterior distribution-the conditional distribution of the target given what is known.
KeywordsPosterior Distribution Monte Carlo Markov Chain Prior Distribution Bayesian Analysis Conditional Distribution
Unable to display preview. Download preview PDF.