A set of independent p-values can be modeled as draws from a mixture of beta distributions. This package provides tools to estmate such a mixture given some data. One component of the mixture is necessarily uniform, representing p-values corresponding to true null hypotheses. The non-uniform components of the mixture represent tests where the alternative hypothesis is true. Using the method described by Allison et al. (2002), p-values are converted to Bayesian posterior probabilities using mixture models and Bayes' rule. This Bayesian posterior probability can be interpreted as the probability that the alternative hypothesis is true. These probabilities can be combined using the standard rules of probability to perform composite hypothesis tests, as described by Erikson, Kim, and Allison (2010).
|Package repository||View on GitHub|
Install the latest version of this package by entering the following in R:
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.