The Choice of sample size; a fully bayesian approach

Current practice of sample size computations is largely based on frequentist or classical methods. In the Bayesian approach the prior information on the unknown parameters is taken into account. In this work we consider a fully Bayesian approach to the sample size determination problem which was introduced by Grundy ef al (1956) and developed by Lindely (1997). This approach treats the problem as a decision problem and employs utility function to find the optimal sample size of a trial. Furthermore, we assume that a regulator authority, who is deciding on whether or not to grant a license to a new service or a new treatment, uses a frequentist approach. We then find the optimal sample size of the trial by maximizing the expected net benefit function which is the expected benefit of subsequent use of the new treatment minus the cost of the trial.

Keywords:

Sample Size Determination , Fully Bayesian Approach , Regulator Authority , Expected Net Benefit , Cost , Normal Distribution , Number of Subsequent Users