SEQBIN
Bayesian/Frequentist Designs for Sequential/Group-Sequential
Trials of a Single Binomial Parameter
Description
This program creates Bayesian designs for either a sequential or
group-sequential single-sample binomial trial. In particular, let
theta be the probability of the event of interest (which in clinical
trials could be either toxicity or response). The trial terminates
when there is sufficient evidence that the chosen one of the three
conditions holds, where ph0 is a user-specified null hypothesis value
of theta. A Bayesian design method is used: the designer specifies a
beta prior distribution on theta; the trial ceases when the posterior
distribution posits that the probability that theta is greater than
(or less than depending on the direction of the test) ph0 is
sufficiently large. For a frequentist, the value of sufficiently
large can be calculated to provide (close to) a specified significance
level.
Contact: Barry W. Brown
Language: Fortran 95
