Model selection and estimation in the Gaussian graphical model
The DOI matches its Crossref record; 16 of the 16 references it cites were found in a registry or an index.
Abstract
We propose penalized likelihood methods for estimating the concentration matrix in the Gaussian graphical model. The methods lead to a sparse and shrinkage estimator of the concentration matrix that is positive definite, and thus conduct model selection and estimation simultaneously. The implementation of the methods is nontrivial because of the positive definite constraint on the concentration matrix, but we show that the computation can be done effectively by taking advantage of the efficient maxdet algorithm developed in convex optimization. We propose a BIC-type criterion for the selection of the tuning parameter in the penalized likelihood methods. The connection between our methods and existing methods is illustrated. Simulations and real examples demonstrate the competitive performance of the new methods.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“These methods have the advantage to be able to recover the off diagonal structure of the covariance matrix.”
From Signal Estimation and Uncertainties Extraction in Terahertz Time-Domain Spectroscopy (2025) · cites this paper as [58]
Citations
Cites
Boyd, S., & Vandenberghe, L. (2004) · doi:10.1017/cbo9780511804441 · Cited by 56,601
Regression shrinkage and selection via the lasso
Tibshirani, R. (1996) · Journal of the Royal Statistical Society Series B: Statistical Methodology · doi:10.1111/j.2517-6161.1996.tb02080.x · Cited by 51,482 · cited 1×
“This is similar to the idea of the lasso in linear regression (Tibshirani, 1996).”
MethodGraphical Models in Applied Multivariate Statistics
Goodall, C. R. (1991) · Technometrics · doi:10.1080/00401706.1991.10484880 · Cited by 1,849
Asymptotics for lasso-type estimators
Fu, W., & Knight, K. (2000) · The Annals of Statistics · doi:10.1214/aos/1015957397 · Cited by 1,276 · cited 2×
“, X(p)) be a p-dimensional random vector following a multivariate normal distribution Np(µ, ) with unknown mean µ and nonsingular covariance matrix .”
BackgroundMethodInfluentialHeuristics of instability and stabilization in model selection
Breiman, L. (1996) · The Annals of Statistics · doi:10.1214/aos/1032181158 · Cited by 1,146 · cited 1×
“As demonstrated by Breiman (1996), the discrete nature of such procedures often leads to instability of the estimator: small changes in the data may result in very different estimates.”
BackgroundBetter Subset Regression Using the Nonnegative Garrote
Breiman, L. (1995) · Technometrics · doi:10.1080/00401706.1995.10484371 · Cited by 1,043 · cited 2×
“As demonstrated by Breiman (1996), the discrete nature of such procedures often leads to instability of the estimator: small changes in the data may result in very different estimates.”
BackgroundMethodIntroduction to Graphical Modelling
Edwards, D. (2000) · Springer Texts in Statistics · doi:10.1007/978-1-4612-0493-0 · Cited by 825
Determinant Maximization with Linear Matrix Inequality Constraints
Vandenberghe, L., Boyd, S., & Wu, S.-P. (1998) · SIAM Journal on Matrix Analysis and Applications · doi:10.1137/s0895479896303430 · Cited by 652 · cited 2×
“To use the algorithm of Vandenberghe et al. (1998), it is also required that Fi, i = 1,. .”
MethodMultivariate Dependencies: Models, Analysis and Interpretation.
Murphy, S. A., Cox, D. R., & Wermouth, N. (1998) · Biometrics · doi:10.2307/2534028 · Cited by 487
Sparsistency and rates of convergence in large covariance matrix estimation
Lam, C., & Fan, J. (2009) · The Annals of Statistics · doi:10.1214/09-aos720 · Cited by 418
Cited by
Items 601–700 of 1,000
Chen, J. (2020) · Psychometrika · doi:10.1007/s11336-020-09724-3 · Cited by 12
Inference on Multi-level Partial Correlations Based on Multi-subject Time Series Data
Qiu, Y., & Zhou, X. (2021) · Journal of the American Statistical Association · doi:10.1080/01621459.2021.1917417 · Cited by 14
“Graphical lasso (Yuan and Lin 2007; Friedman, Hastie, and Tibshirani 2008) and neighborhood selection (Meinshausen and Bühlmann 2006) are two typical (cid:2) 1 penalization methods.”
BackgroundBayesian Topology Learning and noise removal from network data
Ramezani-Mayiami, M., Hajimirsadeghi, M., Skretting, K., Dong, X., Blum, R. S., & Poor, H. V. (2021) · Discover Internet of Things · doi:10.1007/s43926-021-00011-w · Cited by 12
Monitoring Heterogeneous Multivariate Profiles Based on Heterogeneous Graphical Model
Wu, H., Zhang, C., & Li, Y. (2021) · Technometrics · doi:10.1080/00401706.2021.1935325 · Cited by 12
“One way is to optimize the graphical lasso criterion, essentially a Gaussian log-likelihood with the addition of a lasso type penalty on the entries of the precision matrix (Taylor, Worsley, and Gosselin 2007).”
MethodStatistical Methods for Microbiome Compositional Data Network Inference: A Survey
Chen, L., Wan, H. S., He, Q., He, S., & Deng, M. (2022) · Journal of Computational Biology · doi:10.1089/cmb.2021.0406 · Cited by 12
Discussion: Latent variable graphical model selection via convex optimization
Yuan, M. (2012) · The Annals of Statistics · doi:10.1214/12-aos979 · Cited by 12
Self-citationAn efficient method for identifying statistical interactors in gene association networks
Andrei, A., & Kendziorski, C. M. (2009) · Biostatistics · doi:10.1093/biostatistics/kxp025 · Cited by 17
“Several approaches are now available for estimating the paCor matrix for sparse graphs when n << p (Schäfer and Strimmer, 2005a, 2005b; Li and Gui, 2006; Meinshausen and Bühlman, 2006; Yuan and Lin, 2007; Friedman and others, 2008).”
MethodA focused information criterion for graphical models in fMRI connectivity with high-dimensional data
Pircalabelu, E., Claeskens, G., Jahfari, S., & Waldorp, L. (2015) · The Annals of Applied Statistics · doi:10.1214/15-aoas882 · Cited by 12
“In principle, any type of penalty such as an (cid:96) 1 (Meinshausen and B¨uhlmann, 2006) or bridge (Fu, 1998), elastic net (Zou and Hastie, 2005), adaptive lasso (Zou, 2006), SCAD (Fan and Li, 2001), etc. can be used, see below.”
BackgroundMultiple Gaussian graphical estimation with jointly sparse penalty
Tao, Q., Huang, X., Wang, S., Xi, X., & Li, L. (2016) · Signal Processing · doi:10.1016/j.sigpro.2016.03.009 · Cited by 12
A Statistical Graphical Model of the California Reservoir System
Taeb, A., Reager, J. T., Turmon, M., & Chandrasekaran, V. (2017) · Water Resources Research · doi:10.1002/2017wr020412 · Cited by 14
“Recent work (Friedman et al., 2008; Yuan & Lin, 2007) has identified a way around this road block by using a convex relaxation: Ĥ5 arg min H2S55 2‘ðH;DtrainÞ1k jjHjj1; s:t: H 0 : (2) The notation S55 denotes the set of symmetric 55 3 55 matrices.”
BackgroundGraphical Model Selection for Gaussian Conditional Random Fields in the Presence of Latent Variables
Frot, B., Jostins, L., & McVean, G. T. (2018) · Journal of the American Statistical Association · doi:10.1080/01621459.2018.1434531 · Cited by 16
“In practice, model selection in the context of GGMs is often performed using 1-regularized maximum likelihood estimators (MLEs) such as the ones introduced by Banerjee, El Ghaoui and d’Aspremont (2008); Yuan and Lin (2007), and the socalled graphical lasso (Friedman, Hastie, and Tibshirani 2008).”
MethodLinear-Time Algorithm for Learning Large-Scale Sparse Graphical Models
Fattahi, S., Zhang, R. Y., & Sojoudi, S. (2019) · IEEE Access · doi:10.1109/access.2018.2890583 · Cited by 12
Maximum regularized likelihood estimators: A general prediction theory and applications
Zhuang, R., & Lederer, J. (2018) · Stat · doi:10.1002/sta4.186 · Cited by 12
“Recall that the graphical lasso (Yuan & Lin, 2007; Friedman et al., 2008) is formulated as bƒ 2 argmin ƒ2Sp CC ´ 1 n nX iD1 tr Xi.”
MethodInfluentialJoint learning of multiple gene networks from single-cell gene expression data
Wu, N., Fu, Y., Ou-Yang, L., Zhu, Z., & Xie, W. (2020) · Computational and Structural Biotechnology Journal · doi:10.1016/j.csbj.2020.09.004 · Cited by 12
“Based on the assumption that gene networks are sparse, Yuan and Lin [23] imposed a ‘ 1-norm penalty on the non-diagonal elements of the precision matrix when estimating the precision matrix.”
MethodConic optimization for control, energy systems, and machine learning: Applications and algorithms
Zhang, R. Y., Josz, C., & Sojoudi, S. (2018) · Annual Reviews in Control · doi:10.1016/j.arcontrol.2018.11.002 · Cited by 12
“Graphical lasso (GL) is a popular method for estimating the inverse covariance matrix [47]–[49].”
MethodA Joint Fairness Model with Applications to Risk Predictions for Underrepresented Populations
Do, H., Nandi, S., Putzel, P., Smyth, P., & Zhong, J. (2022) · Biometrics · doi:10.1111/biom.13632 · Cited by 13
“Pe-nalized log-likelihood approaches have often been used by other authors like Yuan and Lin (2007), Friedman et al. (2007b) etc. for similar estimation purposes while minimizing the disparities in estimates across groups.”
MethodVariable Selection in Nonparametric and Semiparametric Regression Models
Racine, J. S., Su, L., Ullah, A., Su, L., & Zhang, Y. (2014) · Oxford University Press eBooks · doi:10.1093/oxfordhb/9780199857944.013.009 · Cited by 12
“…Lasso (e.g., Wang and Leng, 2008; Wei and Huang, 2010), fused Lasso (e.g., Tibshirani et al., 2005; Rinaldao, 2009), the graphical Lasso (e.g., Yuan and Lin, 2007; Friedman et al., 2008), the Dantzig selector (e.g., Candès and Tao, 2007), near isotonic regularization (e.g., Tibshirani et al.,…”
BackgroundOn estimation of the diagonal elements of a sparse precision matrix
Balmand, S., & Dalalyan, A. S. (2016) · Electronic Journal of Statistics · doi:10.1214/16-ejs1148 · Cited by 16
SMURC: High-Dimension Small-Sample Multivariate Regression With Covariance Estimation
Bayar, B., Bouaynaya, N. C., & Shterenberg, R. (2016) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2016.2515993 · Cited by 12
Huang, S., Jin, J., & Yao, Z. (2016) · The Annals of Statistics · doi:10.1214/15-aos1392 · Cited by 16
“The glasso [43] is a well-known approach which estimates by optimizing the 1-penalized objective function of the log-likelihood associated with ̂.”
MethodDong, K., & Zhang, S. (2020) · Briefings in Bioinformatics · doi:10.1093/bib/bbaa120 · Cited by 13
“The third term of this equation is the group lasso penalty [40] across all K inverse covariance matrices.”
BackgroundBayesian clustering in decomposable graphs
Bornn, L., & Caron, F. (2011) · Bayesian Analysis · doi:10.1214/11-ba630 · Cited by 15
“It is possible using optimization methods to find the graph which best fits the data according to some metric [23, 30, 13]; alternatively Bayesian model averaging may be used to accommodate for uncertainty in the estimated graph, or maximum a posteriori estimation may be used to select a given model…”
MethodAdaptive estimation of covariance matrices via Cholesky decomposition
Verzélen, N. (2010) · Electronic Journal of Statistics · doi:10.1214/10-ejs580 · Cited by 12
“If there is not such an ordering, d’Aspremont et al. [3] and Yuan and Lin [33] have adopted a penalized likelihood approach by applying a l 1 penalty to the entries of the precision matrix.”
BackgroundMethodInfluentialFull Covariance Modelling for Speech Recognition
Bell, P. (2010) · ERA · Cited by 13
Network inference in matrix-variate Gaussian models with non-independent noisePreprint
Dahl, A., Hore, V., Iotchkova, V., & Marchini, J. (2013) · arXiv (Cornell University) · doi:10.48550/arxiv.1312.1622 · Cited by 12
“In this case a sparse C can be learned using the Graphical Lasso (Glasso), which puts an `1 penalty on C and results in inferences of a Gaussian Graphical Model (GGM) [1, 13].”
MethodTopology Estimation in Bulk Power Grids: Guarantees on Exact RecoveryPreprint
Deka, D., Talukdar, S., Chertkov, M., & Salapaka, M. V. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1707.01596 · Cited by 20
“We use DC-PF and LC-PF models to generate i.i.d. samples of nodal voltages and use them to estimate the inverse covariance matrix for the graphical model through Graphical Lasso.”
MethodThresholded Adaptive Validation: Tuning the Graphical Lasso for Graph RecoveryPreprint
Laszkiewicz, M., Fischer, A., & Lederer, J. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2005.00466 · Cited by 14
“The probably most-frequently used sparsity encouraging estimation procedure for Gaussian graphical models is the graphical lasso (Yuan & Lin, 2007) Θ̂r = argmin Ω∈S+d { tr [ 1 n n∑ i=1 ( z(i) )> z(i)Ω ] − log [det[Ω]] + r‖Ω‖1,off } , (2) where S+d is the set of positive definite and symmetric…”
BackgroundMethodInfluentialSparse High Dimensional Models in EconomicsPreprint
Fan, J., Lv, J., & Lei, Q. (2010) · SSRN Electronic Journal · doi:10.2139/ssrn.1659322 · Cited by 12
LASSO-Driven Inference in Time and SpacePreprint
Chernozhukov, V., Härdle, W. K., Huang, C., & Wang, W. (2018) · SSRN Electronic Journal · doi:10.2139/ssrn.3188362 · Cited by 14
A Nodewise Regression Approach to Estimating Large Portfolios
Callot, L., Caner, M., Önder, A. Ö., & Ulaşan, E. (2019) · Figshare · doi:10.6084/m9.figshare.10022810 · Cited by 12
CARE: Large Precision Matrix Estimation for Compositional Data
Zhang, S., Wang, H., & Lin, W. C. (2024) · Journal of the American Statistical Association · doi:10.1080/01621459.2024.2335586 · Cited by 18
“Instead, the likelihood-based approach estimates the precision matrix as a whole by directly penalizing the joint Gaussian likelihood with the graphical lasso (Yuan and Lin 2007; Fried-man, Hastie, and Tibshirani 2008) or nonconvex penalties (Lam and Fan 2009).”
Oyegoke, O. A., Adekeye, K. S., Olaomi, J. O., & Malela‐Majika, J. (2024) · Quality and Reliability Engineering International · doi:10.1002/qre.3687 · Cited by 11
Spatial-Sign based Robust Sparse Precision Matrix Estimation and its Applications
Lu, Z., & Feng, L. (2026) · Journal of Computational and Graphical Statistics · doi:10.1080/10618600.2026.2743036 · Cited by 11
Inverse covariance estimation from data with missing values using the Concave-Convex Procedure
Thai, J., Hunter, T., Akametalu, A. K., Tomlin, C. J., & Bayen, A. M. (2014) · IEEE Conference on Decision and Control (CDC) · doi:10.1109/cdc.2014.7040287 · Cited by 12
Graphical lasso quadratic discriminant function and its application to character recognition
Xu, B., Huang, K., King, I., Liu, C., Sun, J., & Naoi, S. (2013) · Neurocomputing · doi:10.1016/j.neucom.2012.08.073 · Cited by 11
Randomized Block Proximal Damped Newton Method for Composite Self-Concordant Minimization
Lu, Z. (2017) · SIAM Journal on Optimization · doi:10.1137/16m1082767 · Cited by 13
“, see [39, 6, 9, 17]), which is in the form of: min X∈SN ++ 〈S,X〉 − log det(X) + ρ ∑”
Scherr, S., & Zhou, J. (2019) · Communication Methods and Measures · doi:10.1080/19312458.2019.1677882 · Cited by 11
On variable ordination of Cholesky‐based estimation for a sparse covariance matrix
Kang, X., & Deng, X. (2020) · Canadian Journal of Statistics · doi:10.1002/cjs.11564 · Cited by 11
“Their idea is similar to the graphical Lasso for inverse covariance matrix estimation found in the literature (Yuan & Lin, 2007; Friedman, Hastie & Tibshirani, 2008; Rocha, Zhao & Yu, 2008; Rothman et al., 2008; Yuan, 2008, 2010; Deng & Yuan, 2009), but the computation is much more complicated due…”
MethodFunctional Bayesian Networks for Discovering Causality from Multivariate Functional Data
Zhou, F., He, K., Wang, K., Xu, Y., & Ni, Y. (2023) · Biometrics · doi:10.1111/biom.13922 · Cited by 11
“While graphical models have been extensively studied for vector-and matrix-variate data (Yuan and Lin, 2007; Wang and West, 2009; Leng and Tang, 2012; Ni et al., 2017), only recently have there been several developments for the functional data.”
BackgroundOn lq estimation of sparse inverse covariance
Marjanovic, G., & Hero, A. O. (2014) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2014.6854322 · Cited by 12
“Index Terms— sparsity, lq penalty, non-convex, precision matrix, optimality conditions.”
BackgroundMethodRobust and Sparse Estimation of the Inverse Covariance Matrix Using Rank Correlation Measures
Croux, C., & Oellerer, V. (2016) · Springer eBooks · doi:10.1007/978-81-322-3643-6_3 · Cited by 15
Graph learning with Laplacian constraints: Modeling attractive Gaussian Markov random fields
Egilmez, H. E., Pavéz, E., & Ortega, A. (2016) · Asilomar Conference on Signals, Systems and Computers (ACSSC) · doi:10.1109/acssc.2016.7869621 · Cited by 11
“Some algorithmic extensions of the graphical lasso are discussed in [7], and several alternatives are presented in [8]–[11].”
MethodNode-based differential network analysis in genomics
Zhang, X.-F., Ou-Yang, L., & Yan, H. (2017) · Computational Biology and Chemistry · doi:10.1016/j.compbiolchem.2017.03.010 · Cited by 11
Avagyan, V., Alonso, A. M., & Nogales, F. J. (2017) · Journal of Computational and Graphical Statistics · doi:10.1080/10618600.2017.1340890 · Cited by 11
“(8) The KLL function has been used widely in the prior research on estimation of covariance and precision matrices (see, e.g., Yuan and Lin 2007; Rothman et al. 2008; Yin and Li 2013).”
MethodEftekhari, A., Bollhöfer, M., & Schenk, O. (2018) · International Conference for High Performance Computing, Networking, Storage and Analysis (SC) · doi:10.1109/sc.2018.00023 · Cited by 11
“The resulting problem is nonsmooth but convex and, thus, there are a larger number of approaches from convex optimization which have been developed such as (inexact) interior point methods [11]– [13], blockwise descent methods [14]–[17], iterative thresh-”
BackgroundTuning-Free Heterogeneous Inference in Massive Networks
Ren, Z., Kang, Y., Fan, Y., & Lv, J. (2018) · Journal of the American Statistical Association · doi:10.1080/01621459.2018.1537920 · Cited by 15
“Among those endeavors, a majority of the work has focused primarily on the case of a single Gaussian graphical model; see, for example, [26, 39, 13, 11, 37, 6, 29, 21, 40, 30, 12], among many others.”
BackgroundHierarchical Normalized Completely Random Measures for Robust Graphical Modeling
Cremaschi, A., Argiento, R., Shoemaker, K., Peterson, C. B., & Vannucci, M. (2019) · Bayesian Analysis · doi:10.1214/19-ba1153 · Cited by 12
“In particular, the graphical lasso (Meinshausen and Bühlmann, 2006; Yuan and Lin, 2007; Friedman et al., 2008), which imposes an L1 penalty on the sum of the absolute values of the entries of the precision matrix, is a popular method for achieving the desired sparsity.”
MethodSymmetry Classes and Matrix Representations of the 2D Flexoelectric Law
Abdoul-Anziz, H., Auffray, N., & Desmorat, B. (2020) · Symmetry · doi:10.3390/sym12040674 · Cited by 11
Fused Graphical Lasso for Brain Networks with Symmetries
Ranciati, S., Roverato, A., & Luati, A. (2021) · Journal of the Royal Statistical Society Series C (Applied Statistics) · doi:10.1111/rssc.12514 · Cited by 18
“on structure learning for graphical models. In recent years, much interest has focused on the estimation of concentration graph models through the use of ‘ 1 (lasso) regularization. More specically, Yuan and Lin (2007), Banerjee and others (2008) and Friedman and others (2008) proposed the graphical lasso estimator bg…”
MethodBlock-enhanced precision matrix estimation for large-scale datasets
Eftekhari, A., Pasadakis, D., Bollhöfer, M., Scheidegger, S., & Schenk, O. (2021) · Journal of Computer Science · doi:10.1016/j.jocs.2021.101389 · Cited by 14
Tian, C., Jiang, D., Hammer, A. J., Sharpton, T. J., & Jiang, Y. (2023) · Journal of the American Statistical Association · doi:10.1080/01621459.2022.2164287 · Cited by 11
New Exploratory Tools for Extremal Dependence: $$\chi $$ Networks and Annual Extremal Networks
Huang, W. K., Cooley, D., Ebert‐Uphoff, I., Chen, C., & Chatterjee, S. (2019) · Journal of Agricultural Biological and Environmental Statistics · doi:10.1007/s13253-019-00356-4 · Cited by 11
“, 2000) (for their use in climate science, see for example (Ebert-Uphoff and Deng, 2014)); on combining the concept of Granger causality with vector autoregression (VAR) (Lütkepohl, 2007) or LASSO (Tibshirani, 1996) methods, as known as Graphical Lasso or GLasso, (Yuan and Lin, 2007; Friedman et al., 2008) (for their…”
MethodBayesian inference of networks across multiple sample groups and data types
Shaddox, E., Peterson, C. B., Stingo, F. C., Hanania, N. A., Cruickshank-Quinn, C. I., Kechris, K. J., Bowler, R. P., & Vannucci, M. (2018) · Biostatistics · doi:10.1093/biostatistics/kxy078 · Cited by 11
Learning Moral Graphs in Construction of High-Dimensional Bayesian Networks for Mixed Data
Xu, S., Jia, B., & Liang, F. (2019) · Neural Computation · doi:10.1162/neco_a_01190 · Cited by 12
“…of the gaussian distribution, some efficient algorithms have been developed for learning the structure of GGMs—for example, graphical Lasso (Yuan & Lin, 2007; Friedman, Hastie, & Tibshirani, 2008), nodewise regression (Meinshausen & Bühlmann, 2006), and ψ-learning (Liang, Song, & Qiu,…”
InfluentialBayesian graph selection consistency under model misspecification
Niu, Y., Pati, D., & Mallick, B. K. (2020) · Bernoulli · doi:10.3150/20-bej1253 · Cited by 11
“There has been a growing literature on model selection consistency in Gaussian graphical models from a frequentist point of view [35, 31, 46, 16].”
BackgroundAdaptive Variable Clustering in Gaussian Graphical Models
Sun, S., Zhu, Y., & Xu, J. (2014) · International Conference on Artificial Intelligence and Statistics · Cited by 11
Network-Based Discriminant Analysis for Multiclassification
Chen, L. (2022) · Journal of Classification · doi:10.1007/s00357-022-09414-y · Cited by 11
Inferring Block Structure of Graphical Models in Exponential Families
Sun, S., Wang, H., & Xu, J. (2015) · International Conference on Artificial Intelligence and Statistics · Cited by 11
Learning Ising Models with Independent FailuresPreprint
Goel, S., Kane, D. M., & Klivans, A. R. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1902.04728 · Cited by 21
“hical models assuming the underlying graph is a tree. There have been many works for learning Ising models under various assumptions on the structure of the underlying graph (e.g., Lee et al. (2007); Yuan and Lin (2007); Ravikumar et al. (2010); Yang et al. (2012)). For Ising models specifically, the first assumption-fr…”
MethodHigh-dimensional Gaussian graphical model for network-linked dataPreprint
Li, T., Cheng, Q., Levina, E., & Zhu, J. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1907.02443 · Cited by 11
“Many penalized likelihood methods have been proposed as well [Yuan and Lin, 2007, Banerjee et al., 2008, Rothman et al., 2008, d’Aspremont et al., 2008, Friedman et al., 2008].”
MethodNeurometabolic network (NMetNet) for functional neurological disorder in children and adolescents
Lan, Z., Foster, S., Charney, M. F., van Grinsven, M., Breedlove, K. M., Kozlowska, K., & Lin, A. P. (2025) · NeuroImage Clinical · doi:10.1016/j.nicl.2025.103767 · Cited by 10
“…graphical lasso (Banerjee et al., 2008; Danaher et al., 2014; Friedman et al., 2008; Guo et al., 2011; Meinshausen & Bühlmann, 2006; Wang, 2012; Yuan & Lin, 2007; Wang, 2012) further shrinks the close-to-zero entries to exact zeros, producing a more deterministic approach to identifying…”
Convex formulation for regularized estimation of structural equation models
Pruttiakaravanich, A., & Songsiri, J. (2019) · Signal Processing · doi:10.1016/j.sigpro.2019.107237 · Cited by 11
Joint Estimation of the Two-Level Gaussian Graphical Models Across Multiple Classes
Shan, L., Qiao, Z., Cheng, L., & Kim, I. (2019) · Journal of Computational and Graphical Statistics · doi:10.1080/10618600.2019.1694522 · Cited by 10
“Penalized MLEs are estimated by different algorithms: Yuan and Lin (2007) used the determinant maximization (MAXDET) algorithm; Friedman, Hastie, and Tibshirani (2008) used the clockwise coordinate descent approach and developed the graphical lasso (glasso) algorithm that is remarkably fast;…”
BackgroundMethodInfluentialMultivariate regression of mixed responses for evaluation of visualization designs
Kang, X., Chen, X., Jin, R., Wu, H., & Deng, X. (2020) · IISE Transactions · doi:10.1080/24725854.2020.1755068 · Cited by 10
“…X 1 nh trðUTUXÞ log jXj þ k2jjXjj1 : It has the same form as Graphical LASSO (GLASSO), which has been extensively studied in the literature, for example, Yuan and Lin (2007), Friedman et al. (2008), Rocha et al. (2008), Rothman et al. (2008), Deng and Yuan (2009), Lam and Fan (2009), Raskutti…”
Partial correlation graphical LASSO
Carter, J. S., Rossell, D., & Smith, J. Q. (2023) · Scandinavian Journal of Statistics · doi:10.1111/sjos.12675 · Cited by 20
“For example, Yuan and Lin (2007) argued for a GLASSO framework where one does not penalize the diagonal.”
BackgroundMethodEstimating Finite Mixtures of Ordinal Graphical Models
Lee, K. H., Chen, Q., DeSarbo, W. S., & Xue, L. (2021) · Psychometrika · doi:10.1007/s11336-021-09781-2 · Cited by 10
Reproducible learning in large-scale graphical models
Zhou, J., Li, Y., Zheng, Z., & Li, D. (2021) · Journal of Multivariate Analysis · doi:10.1016/j.jmva.2021.104934 · Cited by 10
De-noising analysis of noisy data under mixed graphical models
Chen, L., & Yi, G. Y. (2022) · Electronic Journal of Statistics · doi:10.1214/22-ejs2028 · Cited by 10
Robust Multivariate Lasso Regression with Covariance Estimation
Chang, L., & Welsh, A. H. (2022) · Journal of Computational and Graphical Statistics · doi:10.1080/10618600.2022.2118752 · Cited by 10
Spectral CUSUM for Online Network Structure Change Detection
Zhang, M., Xie, L., & Xie, Y. (2023) · IEEE Transactions on Information Theory · doi:10.1109/tit.2023.3259185 · Cited by 10
“For instance, penalized likelihood method is proposed by [23] for estimating the concentration matrix in the Gaussian graphical”
MethodA Graph-Assisted Framework for Multiple Graph Learning
Zhang, X., & Wang, Q. (2024) · IEEE Transactions on Signal and Information Processing over Networks · doi:10.1109/tsipn.2024.3352236 · Cited by 11
“Digital Object Identifier 10.1109/TSIPN.2024.3352236 studied models, in which a precision matrix representing the graph topology is learned from the observed data [3], [7].”
BackgroundInfluentialContinuous Estimation of Distribution Algorithms Based on Factorized Gaussian Markov Networks
Karshenas, H., Santana, R., Bielza, C., & Larrañaga, P. (2012) · Adaptation, learning, and optimization · doi:10.1007/978-3-642-28900-2_10 · Cited by 10
Joint Skeleton Estimation of Multiple Directed Acyclic Graphs for Heterogeneous Population
Liu, J., Sun, W., & Liu, Y. (2018) · Biometrics · doi:10.1111/biom.12941 · Cited by 10
“The first step is a Gaussian graphical model estimation problem, which is well studied in the literature (Meinshausen and Bühlmann, 2006; Yuan and Lin, 2007).”
Extended lasso-type MARS (LMARS) model in the description of biological network
Ağraz, M., & Purutçuoğlu, V. (2018) · Journal of Statistical Computation and Simulation · doi:10.1080/00949655.2018.1531414 · Cited by 10
“Yuan and Lin [9] propose a penalized likelihood method to infer (cid”
MethodDiNeR: a Differential graphical model for analysis of co-regulation Network Rewiring
Zhang, J., Liu, J. J., Lee, D., Lou, S., Chen, Z., Gürsoy, G., & Gerstein, M. (2020) · BMC Bioinformatics · doi:10.1186/s12859-020-03605-3 · Cited by 11
Network traffic anomaly detection based on sliding window
Jiang, D., Liu, J., Xu, Z., & Qin, W. (2011) · International Conference on Electrical and Control Engineering · doi:10.1109/iceceng.2011.6057677 · Cited by 10
“DPCA is an efficient implementation of distributed PCA based on a graphical model [10].”
MethodDistribution-free tests of mean vectors and covariance matrices for multivariate paired data
Li, E., Lim, J., Kim, K., & Lee, S. (2011) · Metrika · doi:10.1007/s00184-011-0355-7 · Cited by 10
“The tuning parameter in the graphical lasso was chosen such that the BIC-type criterion of Yuan and Lin (2007) was minimized.”
MethodReplicates in high dimensions, with applications to latent variable graphical models
Tan, K. M., Ning, Y., Witten, D., & Liu, H. (2016) · Biometrika · doi:10.1093/biomet/asw050 · Cited by 12
Yu, P. L. H., Wang, X., & Zhu, Y. (2017) · Computational Statistics & Data Analysis · doi:10.1016/j.csda.2017.04.004 · Cited by 10
Multilevel Gaussian graphical model for multilevel networks
Cheng, L., Shan, L., & Kim, I. (2017) · Journal of Statistical Planning and Inference · doi:10.1016/j.jspi.2017.05.003 · Cited by 10
Robust Gaussian Graphical Model Estimation with Arbitrary Corruption
Wang, L., & Gu, Q. (2017) · International Conference on Machine Learning · Cited by 10
Adjusted regularization in latent graphical models: Application to multiple-neuron spike count data
Vinci, G., Ventura, V., Smith, M. A., & Kass, R. E. (2018) · The Annals of Applied Statistics · doi:10.1214/18-aoas1190 · Cited by 15
“…prior parameters are the same for all (i, j), and m ij = 0: for example, Gaussian/ridge (Hoerl and Kennard, 1970), Laplace/lasso (Tibshirani, 1996; Yuan and Lin, 2007; Friedman, Hastie, and Tibshirani, 2008; Rothman et al., 2008; Wang, 2012; Mazumder and Hastie, 2012), spike-and-slab with Laplace…”
BackgroundMethodInfluentialDarabi-Golestan, F., Hezarkhani, A., & Zare, M. R. (2019) · Environmental Monitoring and Assessment · doi:10.1007/s10661-019-7516-y · Cited by 10
Common Reducing Subspace Model and Network Alternation Analysis
Wang, W., Zhang, X., & Li, L. (2019) · Biometrics · doi:10.1111/biom.13099 · Cited by 11
“There have been a large number of proposals of covariance or precision estimation for a single population, for example, Yuan and Lin (2007), Friedman et al .”
BackgroundBayesian regularization of Gaussian graphical models with measurement error
Byrd, M., Nghiem, L. H., & McGee, M. (2020) · Computational Statistics & Data Analysis · doi:10.1016/j.csda.2020.107085 · Cited by 10
“There are many practical issues associated with Gaussian graphical models, such as hyperparameter tuning [24], missing data [13], and repeated trials [22], which practitioners need to adjust for a successful analysis.”
BackgroundA review of Gaussian Markov models for conditional independence
Córdoba, I., Bielza, C., & Larrañaga, P. (2019) · Journal of Statistical Planning and Inference · doi:10.1016/j.jspi.2019.09.008 · Cited by 10
“Therefore, the penalized likelihood approach can be formulated more directly, for λ ≥ 0, as Yuan and Lin (2007a) were the first to pursue this approach, and they chose 1 , that is, the off-diagonal elements in Ω , which determine the edges of the resulting undirected graph, are penalized.”
MethodTime-Varying Differential Network Analysis for Revealing Network Rewiring over Cancer Progression
Xu, T., Ou-Yang, L., Yan, H., & Zhang, X.-F. (2019) · IEEE Transactions on Computational Biology and Bioinformatics · doi:10.1109/tcbb.2019.2949039 · Cited by 10
“Gaussian Graphical Models (GGMs), which assume that the gene expression measurements follow a multivariate Gaussian distribution, have been used widely for estimating gene regulatory networks [3], [4], [5].”
MethodRobust Scatter Matrix Estimation for High Dimensional Distributions With Heavy Tail
Lu, J., Han, F., & Liu, H. (2021) · IEEE Transactions on Information Theory · doi:10.1109/tit.2021.3088381 · Cited by 10
“Precision matrix is closely related to graphical models and has been widely studied in the literature: (59) propose a neighborhood pursuit method; (60), (61), (62), and (63) apply penalized likelihood methods to obtain the estimators; (64) and (40) propose graphical Dantzig selector and CLIME estimator; (10) and (9; 1…”
MethodRecent Developments in Model-Based Clustering with Applications
Melnykov, V., Michael, S., & Melnykov, I. V. (2014) · Springer eBooks · doi:10.1007/978-3-319-09259-1_1 · Cited by 13
Graphical model selection with latent variables
Wu, C., Zhao, H., Fang, H., & Deng, M. (2017) · Electronic Journal of Statistics · doi:10.1214/17-ejs1331 · Cited by 14
“1 penalty is often applied to induce sparsity (Yuan and Lin, 2007; Friedman et al., 2008; Rothman et al., 2008; Cai et al., 2011; Zhang and Zou, 2014).”
BackgroundTuning parameter selection for penalized likelihood estimation of inverse covariance matrixPreprint
Gao, X., Pu, D. Q., Wu, Y., & Xu, H. (2009) · arXiv (Cornell University) · doi:10.48550/arxiv.0909.0934 · Cited by 10
“Yuan and Lin (2007) proposed penalized likelihood methods for estimating the concentration matrix with L1 penalty (LASSO) (Tibshirani, 1996).”
BackgroundMethodInfluentialThe Gaussian Graphical Model in Cross-Sectional and Time-Series Data
Epskamp, S., Waldorp, L., Mõttus, R., & Borsboom, D. (2018) · Figshare · doi:10.6084/m9.figshare.6144422 · Cited by 10
Recovering social networks from panel data: identification, simulations and an application
de Paula, Á., Imran Rasul, A., & Souza, P. (2022) · doi:10.1920/wp.cem.2018.5818 · Cited by 10
A Note on Ising Network Analysis with Missing Data
Zhang, S., & Chen, Y. (2024) · Psychometrika · doi:10.1007/s11336-024-09985-2 · Cited by 12
“In terms of computation, we propose a joint Pólya–Gamma augmentation procedure by extending the Pólya–Gamma augmentation procedure for logistic regression (Polson et al., 2013) .”
Vector ℓ0 Sparse Conditional Independence Graphs
Marjanovic, G., & Solo, V. (2018) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2018.8461742 · Cited by 9
“[21, 22, 23, 24, 25, 13, 14, 26, 15] used when computing the oracle precision matrix estimator.”
MethodDifferential network inference via the fused D-trace loss with cross variables
Wu, Y., Li, T., Liu, X., & Chen, L. (2020) · Electronic Journal of Statistics · doi:10.1214/20-ejs1691 · Cited by 9
Yuan, X., Yu, W., Yin, Z., & Wang, G. (2020) · IEEE Access · doi:10.1109/access.2020.3031192 · Cited by 9
“Among them, GGMs are the most popular [37].”
BackgroundData-Driven and Knowledge-Based Algorithms for Gene Network Reconstruction on High-Dimensional Data
Abbaszadeh, O., Azarpeyvand, A., Khanteymoori, A., & Bahari, A. (2020) · IEEE Transactions on Computational Biology and Bioinformatics · doi:10.1109/tcbb.2020.3034861 · Cited by 9
Robust estimators for nondecomposable elliptical graphical models
Vogel, D., & Tyler, D. E. (2014) · Biometrika · doi:10.1093/biomet/asu041 · Cited by 9
Topology Constraints in Graphical Models
Fiori, M., Musé, P., & Sapiro, G. (2012) · neural information processing systems · Cited by 9
Metadata
- Authors
- Yuan, M., & Lin, Y.
- Affiliations
- Georgia Institute of Technology (US)
- University of Wisconsin–Madison (US)
- Venue
- Biometrika
- Publisher
- Oxford University Press (OUP)
- Published
- 2007-02-07
- Volume
- 94
- Issue
- 1
- Pages
- 19-35
- ISSN
- 0006-3444 (ISSN-L)
- 1464-3510
- OpenAlex
- W2081746825
- Semantic Scholar
- 20ee888ca9c3bae784f266405b4d6395c6f32955
- Cited by
- 1,627
- References
- 16
- Citations by year
- 2026: 43 · 2025: 66 · 2024: 60 · 2023: 64 · 2022: 86 · 2021: 131 · 2020: 112 · 2019: 147 · 2018: 142 · 2017: 91 · 2016: 123 · 2015: 98 · 2014: 109 · 2013: 86 · 2012: 85
- FWCI
- 38.99
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- Biometrika (publishedVersion)
- RePEc: Research Papers in Economics (submittedVersion)
- HighWire Press Open Archive (submittedVersion)
- Funders
- National Science Foundation
- Topics
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods and Inference
- Physical Sciences › Mathematics › Statistics and Probability › Random Matrices and Applications
- Physical Sciences › Computer Science › Artificial Intelligence › Bayesian Modeling and Causal Inference
- Keywords
- Gaussian graphical models, model selection, penalized likelihood, sparse estimation, shrinkage estimation, hyperparameter optimization, covariance structure
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.