Bayesian semiparametric clustering of functional predictors

This article proposes a new method for the joint clustering of functional predictors with some outcome of interest. A multivariate adaptive spline model is used to describe the functions, and the outcome is modeled through a generalized linear model with a random intercept. Through specifying the random intercept to follow a Dirichlet process jointly with the random spline coefficients, we obtain a procedure that clusters trajectories according to shape and according to the parameters of the outcome model for each cluster. This very flexible method allows for the incorporation of covariates in the models for both the outcome and the trajectory. We apply the method to post-ovulatory progesterone data from the Early Pregnancy Study and find that the model successfully separates clinical pregnancies from early pregnancy losses.