Sebastian Ratzenböck made the suggestion to let the tool hierarchically estimate the selection probabilities. The idea is to start out with large counting bins and a flat prior for the binomial distribution parameters and then let the resulting posterior be the prior for a refined set of counting bins.
Sebastian Ratzenböck made the suggestion to let the tool hierarchically estimate the selection probabilities. The idea is to start out with large counting bins and a flat prior for the binomial distribution parameters and then let the resulting posterior be the prior for a refined set of counting bins.