February, 2003
© 2000, 2001, 2003 Scott S. Emerson, M.D., Ph.D.
In either case, we make probability statements to quantify
the strength of evidence
Bayesian inference:
·Credible interval covers 100(1- a )% of posterior
distribution of q
·Posterior probability of hypotheses
Frequentist inference:
·Confidence
interval defines hypotheses for which observed data is in
central 100(1- a
)% of sampling
distribution
·P value is probability under the null of observing
as extreme results as observed data