Maximum Likelihood Estimation of Mixture Densities for Binned and Truncated Multivariate Data
Machine Learning - Special issue: Unsupervised learning
Bayesian Core: A Practical Approach to Computational Bayesian Statistics
Bayesian Core: A Practical Approach to Computational Bayesian Statistics
Monte Carlo Statistical Methods
Monte Carlo Statistical Methods
| Hi-index | 0.03 |
The usual Gibbs sampling framework of the Bayesian mixture model is extended to account for binned data. This model involves the addition of a latent variable in the model which represents simulated values from the believed true distribution at each iteration of the algorithm. The technique results in better model fit and recognition of the more subtle aspects of the density of the data.