Publication

Significance Tests for Gaussian Graphical Models Based on Shrunken Densities

Bernal Arzola, V., Guryev, V., Bischoff, R., Horvatovich, P. & Grzegorczyk, M., 20-Jul-2018, proceedings of the 33rd Inter- national Workshop on Statistical Modelling (IWSM), University of Bristol, UK, 16-20 July 2018. University of Bristol, Vol. 2. p. 27 32 p.

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

Copy link to clipboard

Documents

  • Significance Tests for Gaussian Graphical Models Based on Shrunken Densities

    Final author's version, 512 KB, PDF document

    Request copy

  • Proceedings of the 33rd International Workshop on Statistical Modelling Volume I

    19.4 MB, PDF document

    Request copy

Gaussian Graphical Models (GGMs) are important probabilistic graphical models in Statistics. Inferring a GGM’s structure from data implies computing the inverse of the covariance matrix (i.e. the precision matrix). When the number of variables p is larger than the sample size n, the (sample) covariance estimator is not invertible and therefore another estimator is required. Covariance estimators based on shrinkage are more stable (and invertible), however, classical hypothesis testing for the ”shrunk” coefficients is an open challenge. In this paper, we present an exact null-density that naturally includes the shrinkage, and allows an accurate parametric significance test that is accurate and computationally efficient.
Original languageEnglish
Title of host publication proceedings of the 33rd Inter- national Workshop on Statistical Modelling (IWSM), University of Bristol, UK, 16-20 July 2018
PublisherUniversity of Bristol
Pages27
Number of pages32
Volume2
Publication statusPublished - 20-Jul-2018
Event33rd International Workshop on Statistical Modelling - University of Bath, Bristol, United Kingdom
Duration: 16-Jul-201820-Jul-2018
https://people.maths.bris.ac.uk/~sw15190/IWSM2018/IWSM33-2.pdf

Conference

Conference33rd International Workshop on Statistical Modelling
Abbreviated titleIWSM 2018
CountryUnited Kingdom
CityBristol
Period16/07/201820/07/2018
Internet address

Event

33rd International Workshop on Statistical Modelling

16/07/201820/07/2018

Bristol, United Kingdom

Event: Conference

ID: 63162271