11.3 Summary

In this chapter, we introduce the most common Bayesian methods for inference in non-parametric and semi-parametric models. Finite Gaussian and Dirichlet process mixtures, as well as splines, are highly flexible methods; however, they also have limitations, such as the label-switching issue in mixtures and potential multicollinearity in splines. Additionally, it is wise to elicit informative priors and conduct sensitivity analyses to assess the robustness of the results in real-world applications.