joint graphical lasso for inverse covariance estimation across multiple classes

We consider the problem of estimating multiple related Gaussian graphical models from a high dimensional data set with observations belonging to distinct classes. We propose the joint graphical lasso, which borrows strength across the classes to estimate multiple graphical models that share certain...

Full description

Saved in:
Bibliographic Details
Published in:Journal of the Royal Statistical Society. Series B, Statistical methodology Vol. 76; no. 2; pp. 373 - 397
Main Authors: Danaher, Patrick, Wang, Pei, Witten, Daniela M
Format: Journal Article
Language:English
Published: England Blackwell Publishers 01.03.2014
Blackwell Publishing Ltd
John Wiley & Sons Ltd
Oxford University Press
Subjects:
Algorithms
Alternating directions method of multipliers
Analysis of covariance
Covariance
data collection
Gaussian graphical model
Generalization
Generalized fused lasso
Graph theory
Graphical lasso
Graphical models
Group lasso
High dimensional data
Multiplier
multipliers
Network estimation
Optimization
Penalties
Probability
Statistical methods
Statistics
Studies
high-dimensional
generalized fused lasso
Gaussian graphical model
alternating directions method of multipliers
graphical lasso
group lasso
network estimation
ISSN:1369-7412, 1467-9868
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We consider the problem of estimating multiple related Gaussian graphical models from a high dimensional data set with observations belonging to distinct classes. We propose the joint graphical lasso, which borrows strength across the classes to estimate multiple graphical models that share certain characteristics, such as the locations or weights of non‐zero edges. Our approach is based on maximizing a penalized log‐likelihood. We employ generalized fused lasso or group lasso penalties and implement a fast alternating directions method of multipliers algorithm to solve the corresponding convex optimization problems. The performance of the method proposed is illustrated through simulated and real data examples.
Bibliography:http://dx.doi.org/10.1111/rssb.12033
istex:0BDA33974C3C9B708B7C4B7942FC7DEE8E50B850
'Benchmarking the two MCMC strategies for sampling the Bayesian bridge posterior distiution' and 'An empirical study of mixing rates in parameter-expanded Gibbs samplers for sparse Bayesian regression models'
ArticleID:RSSB12033
National Institutes of Health - No. 1R01GM082802; No. P01CA53996; No. U24CA086368; No. DP5OD009145
ark:/67375/WNG-T9FSVP5K-1
SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 14
ObjectType-Article-2
content type line 23
ObjectType-Article-1
ObjectType-Feature-2
ISSN:1369-7412
1467-9868
DOI:10.1111/rssb.12033