Parameter Estimation For Multivariate Generalized Gaussian Distributions

@article{Pascal2013ParameterEF,
  title={Parameter Estimation For Multivariate Generalized Gaussian Distributions},
  author={Fr{\'e}d{\'e}ric Pascal and Lionel Bombrun and Jean-Yves Tourneret and Yannick Berthoumieu},
  journal={IEEE Transactions on Signal Processing},
  year={2013},
  volume={61},
  pages={5960-5971}
}
Due to its heavy-tailed and fully parametric form, the multivariate generalized Gaussian distribution (MGGD) has been receiving much attention in signal and image processing applications. Considering the estimation issue of the MGGD parameters, the main contribution of this paper is to prove that the maximum likelihood estimator (MLE) of the scatter matrix exists and is unique up to a scalar factor, for a given shape parameter β ∈ (0,1). Moreover, an estimation algorithm based on a Newton… 
Globally Convergent Algorithms for Learning Multivariate Generalized Gaussian Distributions
TLDR
Global convergent algorithms based on the block majorization minimization method are proposed to jointly learn all the model parameters in the maximum likelihood estimation setting and the shape parameter is proved to be strictly convex, which to best knowledge is the first result of this kind in the literature.
An efficient multivariate generalized Gaussian distribution estimator: Application to IVA
TLDR
This paper proposes an efficient estimation technique based on the Fisher scoring (FS) and demonstrates its successful application to IVA, and quantifies the performance of MGGD parameter estimation using FS and proves the effectiveness of the new IVA algorithm using simulations.
A New Riemannian Averaged Fixed-Point Algorithm for MGGD Parameter Estimation
TLDR
This work proposes a new FP algorithm, Riemannian averaged FP (RA-FP), which can effectively estimate the scatter matrix for any value of the shape parameter and provides significantly improved performance over existing FP and method-of-moments algorithms for the estimation of the scattering matrix.
Online estimation of MGGD: the Riemannian Averaged Natural Gradient method
  • Jialun Zhou, S. Said, Y. Berthoumieu
  • Computer Science
    2019 IEEE 8th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP)
  • 2019
TLDR
A new method for online estimation of MGGD parameters, called the Riemannian Averaged Natural Gradient (RANG) method, which achieves identical performance, to classical maximum-likelihood estimation, for sufficiently large datasets.
On the characterization, generation, and efficient estimation of the complex multivariate GGD
TLDR
This work develops a fixed-point algorithm for the estimation of parameters of the CMGGD that is both rapid in its convergence and accurate for the complete shape parameter range and quantifies performance against other algorithms while varying noncircularity, shape parameter and data dimensionality.
Multivariate Generalized Gaussian Distribution: Convexity and Graphical Models
TLDR
This work considers covariance estimation in the multivariate generalized Gaussian distribution and shows that the optimizations can be formulated as convex minimization as long the MGGD shape parameter is larger than half and the sparsity pattern is chordal.
Joint Estimation of Location and Scatter in Complex Elliptically Symmetric Distributions
TLDR
A joint estimator, relying on the Tyler’s M-estimator of location and the R-estIMator of shape matrix, is proposed and its Mean Squared Error (MSE) performance is compared with the Semiparametric Cramer-Rao Bound.
Shape parameter estimation for generalized-Gaussian-distributed frequency spectra of audio signals
TLDR
An integration of the estimation into an audio codec for an example of its application resulted in the enhancement of the objective and subjective reconstruction quality, which may be useful in other audio signal processing problems.
Robust M-Estimation Based Bayesian Cluster Enumeration for Real Elliptically Symmetric Distributions
TLDR
A robust criterion for data sets with finite sample size is derived, and an asymptotic approximation is provided to reduce the computational cost at large sample sizes, and the algorithms show a significant robustness improvement in comparison to existing methods.
A Robust and Flexible EM Algorithm for Mixtures of Elliptical Distributions with Missing Data
TLDR
Experimental results on synthetic data demonstrate that the proposed EM algorithm is robust to outliers and can be used with non-Gaussian data, and experiments conducted on real-world datasets show that this algorithm is very competitive when compared to other classical imputation methods.
...
...

References

SHOWING 1-10 OF 59 REFERENCES
Parametric generalized Gaussian density estimation
The primary objective of this paper is to compare the large‐sample as well as the small‐sample properties of different methods for estimating the parameters of a three‐parameter generalized Gaussian
Multivariate Generalized Gaussian Distribution: Convexity and Graphical Models
TLDR
This work considers covariance estimation in the multivariate generalized Gaussian distribution and shows that the optimizations can be formulated as convex minimization as long the MGGD shape parameter is larger than half and the sparsity pattern is chordal.
On the Geometry of Multivariate Generalized Gaussian Models
TLDR
The geometry of the zero-mean multivariate generalized Gaussian distribution (MGGD) and the calculation of geodesic distances on the MGGD manifold are concerns and an application to image texture similarity measurement in the wavelet domain is briefly discussed.
Geodesics on the Manifold of Multivariate Generalized Gaussian Distributions with an Application to Multicomponent Texture Discrimination
TLDR
The modeling of the interband correlation significantly improves classification efficiency, while the GD is shown to consistently outperform the KLD as a similarity measure.
Complex Elliptically Symmetric Distributions: Survey, New Results and Applications
TLDR
Applications of CES distributions and the adaptive signal processors based on ML- and M-estimators of the scatter matrix are illustrated in radar detection problems and in array signal processing applications for Direction-of-Arrival estimation and beamforming.
Image thresholding based on the EM algorithm and the generalized Gaussian distribution
A globally convergent and consistent method for estimating the shape parameter of a generalized Gaussian distribution
  • K. Song
  • Mathematics
    IEEE Transactions on Information Theory
  • 2006
TLDR
It is proved via fixed point arguments that, with probability tending to one, the Newton-Raphson algorithm converges to the unique global root of the sample shape estimating equation from any starting point in Thetamin.
A Complex Generalized Gaussian Distribution— Characterization, Generation, and Estimation
TLDR
This correspondence provides a thorough presentation of the complex-valued GGD by constructing the probability density function, defining a procedure for generating random numbers from the complex GGD, and implementing a maximum likelihood estimation (MLE) procedure for the shape and covariance parameters in the complex domain.
Covariance Structure Maximum-Likelihood Estimates in Compound Gaussian Noise: Existence and Algorithm Analysis
TLDR
The derivation is based on some likelihood functions general properties like homogeneity and can be easily adapted to other recursive contexts and shows the convergence of this recursive scheme, ensured whatever the initialization.
Modeling and Inference with υ-Spherical Distributions
Abstract A new class of continuous multivariate distributions on × ∈ ℜ n is proposed. We define these so-called υ-spherical distributions through properties of the density function in a
...
...