1: \begin{abstract}

2: \noindent 

3: This paper adopts a Bayesian nonparametric mixture model where the mixing distribution 

4: belongs to the wide class of normalized homogeneous completely random measures.

5: We propose a truncation method for the mixing distribution by discarding the weights of the unnormalized

6: measure smaller than a threshold. We prove convergence in law 

7: of  our approximation, provide some theoretical properties and characterize

8: its  posterior distribution so that a blocked Gibbs sampler is devised.

9: 

10: The versatility of the approximation is illustrated  by two different applications. In the first  

11: the normalized Bessel random measure, encompassing the Dirichlet process, is introduced;

12: goodness of fit indexes show its  good performances as mixing measure for

13: density estimation.

14: The second describes how to incorporate covariates in the support of the

15: normalized measure, leading to a linear dependent model for regression and 

16: clustering.

17: 

18: \vskip6pt

19: {{\bf \noindent Keywords}: Bayesian nonparametric mixture models

20: \textbullet~ normalized  completely random measure

21: \textbullet~ blocked Gibbs sampler

22: \textbullet~ finite dimensional approximation

23: \textbullet~ a priori truncation method}

24: \end{abstract}

25: