Bayesian Sample Size Determination For Binomial Proportions. Crossref Volume 38 Issue 23. The structure is as follows.
Https Www Jstor Org Stable 2680864 from
A Dirichlet distribution is assumed to describe prior knowledge of the two success probabilities p1 and p2. MLan Lawrence Josephy and David B. Bayesian sample size determination SupposethatweareinterestedinchoosingthesizenofarandomsampleXn X1X2Xn whose joint density function fnθ we use here the notation for continuous random vari-ables without loss of generality depends on an unknown parameter vector θ that we want to estimate.
In Section 2 in the context of hypothesis testing two cases are considered.
A Dirichlet distribution is assumed to describe prior knowledge of the two success probabilities p1 and p2. The parameter of interest is p p1 - p2. Crossref Volume 38 Issue 23. P 1 p 2 p where p is specified the other a test of H 0.
