Sampling methods for the concentration parameter and discrete baseline of the Dirichlet Process

Abstract There are many models in the current statistical literature for making inferences based on samples selected from a finite population. Parametric may be problematic because inference is sensitive to parametric assumptions. The Dirichlet process (DP) prior very flexible and determines complexity of model. It indexed by two hyper-parameters: baseline distribution concentration parameter. We address distinct problems article. Firstly, we review sampling methods parameter, which use continuous distribution. compare three different methods: adaptive rejection method, mixture Gammas method grid method. also propose new ratio uniforms. Secondly, practice, some survey responses known discrete. If adopted as distribution, model misspecified standard invalid. discrete approach DP sample unobserved population both using Polya urn scheme Multinomial applied our Phytophthora data set.