Two Bayesian/frequentist challenges for categorical data analyses

We discuss two challenging scenarios for frequentist and/or Bayesian inference for categorical data. First, for parameter space regions described by order restrictions, frequentist methods are less straightforward than Bayesian methods, especially for interval estimation. Second, for marginal modeling, frequentist inference is currently feasible only in relatively simplistic settings and Bayesian solutions seem to be essentially non-existent. Both areas have substantial scope for future research. Copyright Sapienza Università di Roma 2014