Online Supplement: Principled Default Priors for Bayesian Hypothesis Testing in Binomial GLMs.

This project archives the simulation code, analysis scripts, and pre-computed outputs for the paper "Principled Default Priors for Bayesian Hypothesis Testing in Binomial GLMs: A Matched-Prior Framework with Empirical Evaluation." The paper develops a default prior framework for logistic and probit regression: matching the prior family to the link function yields a uniform prior on the probability scale, which serves as an interpretable anchor for setting the prior scale via prior predictive simulation. The materials here reproduce figures and tables in the manuscript and supplement without re-running the compute-heavy simulations. The analysis code is also maintained at https://github.com/GidonFrischkorn/PriorsBinomial