<<
>>

F Coda

This chapter reviewed the fundamental features of Bayesian statistics and methods, together with their advantages and disadvantages. Probably the most important and problematical aspect of Bayesian statistics is its absolute depend­ence on prior probabilities, and the chapter discussed some of the issues associ­ated with selecting appropriate priors.

We also worked through several kinds of problems to give the gist of Bayesian thinking and reviewed a few of the marked distinctions between frequentist and Bayesian philosophies of science. Given its emphasis on probabilistically secure answers, Bayesian statistics is gener­ally more suitable for applied, rather than basic, science. Nevertheless, we also discussed Bayesian model selection procedures and I argued that a modified Bayesian approach could be incorporated into hypothesis-testing procedures.

Unless noted otherwise, “statistics” throughout this book refers to frequentist statistics. The next two chapters will call on a number of the concepts and issues discussed in Chapters 5 and 6.

<< | >>
Source: Alger Bradley E.. Defense of the Scientific Hypothesis: From Reproducibility Crisis to Big Data. Oxford University Press,2020. — 449 p.. 2020

More on the topic F Coda: