Estimation of Bayesian Sample Size for Binomial Proportions Using Areas P-tolerance with Lowest Posterior Loss

STIMATION of sample size is one of the questions that we often encounter in research and applied designs. Various methods have been proposed to estimate Bayesian sample size. Freedman and Spiegelhalter [11] and Spiegelhalte et al. [12], used Bayesian approach for anticipating hypothesis testing power. Adcock [1], Pham Gia and Turkkan [10] used it for interval estimation based on normal approximations of posterior densities or intervals based on means and posterior variances. One of the goals of estimating sample size is to make inference or decision about uncertain parameter θ. In classic approach which has been summarized by Desu and Raghavarao in [4], the main problem is to find point estimate