Bayesian Cumulative Probability Models for Continuous and Mixed Outcomes
Nathan James, PhD Vanderbilt University School of Medicine
Ordinal cumulative probability models (CPMs) such as the proportional odds regression model are typically used for discrete ordered outcomes, but can accommodate both continuous and mixed discrete/continuous outcomes. Recent papers by Liu et al. and Tian et al. describe the advantages of ordinal CPMs in this setting using non-parametric maximum likelihood estimation (NPMLE), however the extension to Bayesian inference has not been thoroughly explored. We formulate a Bayesian CPM for continuous or mixed outcome data. Bayesian CPMs have advantages over frequentist CPMs with regard to interpretation, flexibility/extendability, and exact inference (within simulation error) when the log-likelihood is not well-approximated by a quadratic function. We explore characteristics of the Bayesian CPM through simulations and a case study using HIV biomarker data.