The Distribution of the Maximum Likelihood Estimator in Invariant Gaussian Graphical Models and its Application to Likelihood Ratio Tests
- The distribution of the maximum likelihood estimator of the covariance matrix in a class of invariant Gaussian graphical models is determined and seen to belong to the class of generalized Riesz distributions. For testing two nested models, the distribution of the likelihood ratio statistic under the null hypothesis is shown to be of Box type, so that accurate approximation techniques are applicable.
Author: | Andreas Käufl |
---|---|
URN: | urn:nbn:de:bvb:384-opus4-21756 |
Frontdoor URL | https://opus.bibliothek.uni-augsburg.de/opus4/2175 |
Series (Serial Number): | Preprints des Instituts für Mathematik der Universität Augsburg (2012-16) |
Type: | Preprint |
Language: | English |
Publishing Institution: | Universität Augsburg |
Release Date: | 2012/12/17 |
Tag: | maximum likelihood estimator; invariant Gaussian graphical model; likelihood ratio test; generalized Riesz distribution; dimension reduction; box approximation |
GND-Keyword: | Maximum-Likelihood-Schätzung; Likelihood-Quotienten-Test; Graphisches Modell; Wahrscheinlichkeitsverteilung |
Institutes: | Mathematisch-Naturwissenschaftlich-Technische Fakultät |
Mathematisch-Naturwissenschaftlich-Technische Fakultät / Institut für Mathematik | |
Dewey Decimal Classification: | 5 Naturwissenschaften und Mathematik / 51 Mathematik / 510 Mathematik |
Licence (German): | Deutsches Urheberrecht |