Variational methods for model comparison have become popular in the neural computing/machine learning literature. In this paper we explore their application to the Bayesian analysis of mixtures of Gaussians. We also consider how the Deviance Information Criterion, or DIC, devised by Spiegelhalter et al. (2002), can be extended to these types of model by exploiting the use of variational approxi...

The performance of interval estimates in a uniform-beta mixture model is evaluated using three computational strategies. Such a model has found use when modeling a distribution of P -values from multiple testing applications. The number of P -values and the closeness of a parameter to the boundary of its space both play a role in the precision of parameter estimates as does the “nearness” of th...

This paper illustrates the application of resampling methods for model diagnostics of fitted finite mixture models. Convenience functions to perform these methods are available in package flexmix. The results of the application to an artificial example and the seizure data set as described in Grün and Leisch (2010) are reproduced.

We consider a finite mixture model with k components and a kernel distribution from a general parametric family. We consider the problem of testing the hypothesis k = 2 against k ≥ 3. In this problem, the likelihood ratio test has a very complicated large sample theory and is difficult to use in practice. We propose a test based on the likelihood ratio statistic where the estimates of the param...

The region-based coding of the volume data is proposed in the present paper. Since the volume data is the three dimensional one, the visualization method is needed to display it. The region segmentation is indispensable to visualize the desired portions only. In the proposed method, the region segmentation result is utilized to code the volume data; the contours and the colors of the regions ar...