Repulsive Mixtures

Details Repulsive Mixtures Repulsive Mixtures

2012-04-24

arxiv.org/abs/1204.5243v1

Statistics

Methodology

Scientific paper

Discrete mixture models are routinely used for density estimation, clustering and as components in more complex hierarchical models. When used in clustering applications, problems can arise in identifying clusters if mixture components are not sufficiently well separated. Low separation can also create problems for density estimation and robust modeling, as redundant components lead to an unnecessarily complex model and associated computational problems. Current Bayesian practice generates component-specific parameters independently from a common prior, which tends to favor components that are close together. As an alternative, we propose to generate components from a repulsive process. We characterize this repulsive prior theoretically and propose a Markov chain Monte Carlo sampling algorithm for posterior computation. The methods are illustrated using simulated data as well as real datasets.

Affiliated with

Also associated with

No associations

Rating

Repulsive Mixtures does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.

If you have personal experience with Repulsive Mixtures, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Repulsive Mixtures will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFWR-SCP-O-518782