February, 2003
© 2000, 2001, 2003 Scott S. Emerson, M.D., Ph.D.
None yet agreed upon, but…
The concept of having confidence in your conclusions exists
in both Bayesian and frequentist inference
·Both
Bayesian credible intervals and frequentist confidence intervals satisfy
frequentist coverage criteria
In either type of inference, it seems reasonable to
use a consistent definition of “rejection” of hypotheses
·Use of credible interval or confidence interval
·Frequentist tests: equal type I and type II errors