= Posterior equality for proportional likelihoods
{title2=$L_i'=hL_i\ \Longrightarrow\ \pi_i'=\pi_i$}
If every model's <likelihood> is multiplied by the same positive factor <independent> of its <statistical parameter>[parameter] or model index, <Bayes' theorem> cancels that factor from the normalized <posterior distribution>[posterior]. Fixed-count Bernoulli sampling and sampling stopped at a specified positive head count give such proportional <likelihoods> for the same total heads and tails. This conclusion uses their actual sampling <likelihoods>; it is not a claim that every selection or stopping rule can be ignored.
Back to article page