Title: Why Bother with Bayes
Speaker: Dr Thomas Louis
Department of Biostatistics, John Hopkins Bloomberg School of Public Health
Expert Statistical Consultant, Center for Drug Evaluation and Research, US FDA
The use of Bayesian designs and analyses in biomedical, environmental, educational, policy and many other contexts has burgeoned, even though its use entails additional overhead and some risk. Consequently, it is evident that statisticians and collaborators are increasingly finding the approach worth the bother. In the context of this escalating incidence, I highlight a subset of the potential advantages of the Bayesian approach along with some caveats. Examples include designs and analyses with required frequentist properties (Bayes for frequentist) and for fully Bayesian goals (Bayes for Bayes), a well-calibrated CI for a binomial parameter, sample size for a confidence interval, (Bayes) empirical Bayes modeling, comparison of two treatments (including a non-inferiority assessment), clinical trial monitoring, multiplicity issues; and addressing non-standard goals including ranking and histogram estimation.
The Bayesian approach is by no means a panacea (indeed, in 1970, Herman Rubin cautioned, “A good Bayesian does better than a non-Bayesian, but a bad Bayesian gets clobbered.") So, valid development and application places additional obligations on the investigative team, and it isn't always worth the effort. However, the investment can pay big dividends, the cost/benefit relation is increasingly attractive, and in many situations `going Bayes' is definitely worth the bother.
Join Skype Meeting: https://meet.lilly.com/natanegara_fanni/6T31HJQ8
Join by phone
+1 (317) 277-1498 (United States) English (United States)
+1 (855) 545-5910 (United States) English (United States)