On the Bumpy Road to the Dominant Mode.

@article{Zhou2010OnTB,
  title={On the Bumpy Road to the Dominant Mode.},
  author={Hua Zhou and K. Lange},
  journal={Scandinavian journal of statistics, theory and applications},
  year={2010},
  volume={37 4},
  pages={
          612-631
        }
}
  • Hua Zhou, K. Lange
  • Published 2010
  • Mathematics, Medicine
  • Scandinavian journal of statistics, theory and applications
Maximum likelihood estimation in many classical statistical problems is beset by multimodality. This article explores several variations of deterministic annealing that tend to avoid inferior modes and find the dominant mode. In Bayesian settings, annealing can be tailored to find the dominant mode of the log posterior. Our annealing algorithms involve essentially trivial changes to existing optimization algorithms built on block relaxation or the EM or MM principle. Our examples include… Expand
Computational aspects of fitting mixture models via the expectation-maximization algorithm
Improved model-based clustering performance using Bayesian initialization averaging
Mixtures of multivariate power exponential distributions.
A quasi-Newton acceleration for high-dimensional optimization algorithms
Mixtures of Shifted AsymmetricLaplace Distributions
The EM Algorithm
Parsimonious Shifted Asymmetric Laplace Mixtures
Multivariate Response and Parsimony for Gaussian Cluster-Weighted Models
...
1
2
3
4
...

References

SHOWING 1-10 OF 36 REFERENCES
Posterior bimodality in the balanced one‐way random‐effects model
A Tutorial on MM Algorithms
A curious likelihood identity for the multivariate t-distribution
Deterministic annealing EM algorithm
Robust Statistical Modeling Using the t Distribution
The MM Alternative to EM
...
1
2
3
4
...