We present a graphical user interface (GUI) for planning the sample size needed to reach a specified target uncertainty in a Bayesian type A uncertainty evaluation of normal or Poisson… Click to show full abstract
We present a graphical user interface (GUI) for planning the sample size needed to reach a specified target uncertainty in a Bayesian type A uncertainty evaluation of normal or Poisson distributed data. To this end we build on a criterion previously introduced by Martin and Elster (2020 Stat. Methods Appl. 1–21) and called the variation of the posterior variance criterion. This criterion includes, and extends, standard Bayesian sample size planning procedures. Guidance is provided for the elicitation of the required prior knowledge in a way that makes the approach easily accessible for metrologists. The GUI also includes a menu that performs the Bayesian inference after the experiment has been carried out.
               
Click one of the above tabs to view related content.