Bayesian testing of a point null hypothesis based on the latent information prior (Q280652)

Summary: Bayesian testing of a point null hypothesis is considered. The null hypothesis is that an observation, \(x\), is distributed according to the normal distribution with a mean of zero and known variance \(\sigma^2\). The alternative hypothesis is that \(x\) is distributed according to a normal distribution with an unknown nonzero mean, \(\mu\), and variance \(\sigma^2\). The testing problem is formulated as a prediction problem. Bayesian testing based on priors constructed by using conditional mutual information is investigated.