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 901–1,000 of 1,000
Multivariate sparse Laplacian shrinkage for joint estimation of two graphical structures
Yang, Y., Xia, S., & Yang, H. (2022) · Computational Statistics & Data Analysis · doi:10.1016/j.csda.2022.107620 · Cited by 5
A robust Glasso approach to portfolio selection in high dimensions
Ding, W., Shu, L., & Gu, X. (2022) · Journal of Empirical Finance · doi:10.1016/j.jempfin.2022.11.003 · Cited by 5
Bertsimas, D., & Koukouvinos, T. (2023) · European Journal of Operational Research · doi:10.1016/j.ejor.2023.11.036 · Cited by 5
Sparse Gaussian graphical models for speech recognition
Bell, P., & King, S. (2007) · Interspeech (USB) · doi:10.21437/interspeech.2007-571 · Cited by 5
“Recent work [6, 5, 12] on Gaussian graphical model structure learning has focused on the use of the l1-norm of the parameter vector as a penalty term.”
BackgroundMaximum Likelihood Estimation Over Directed Acyclic Gaussian Graphs
Yuan, Y., Shen, X., & Pan, W. (2012) · Statistical Analysis and Data Mining The ASA Data Science Journal · doi:10.1002/sam.11168 · Cited by 5
“Therefore, different methods using penalization have been proposed for sparse estimation of graphical models [12, 22, 6, 17].”
MethodGaskins, J., & Daniels, M. J. (2015) · Journal of Computational and Graphical Statistics · doi:10.1080/10618600.2015.1028549 · Cited by 5
“For a single group with no partitioning, our prior can be considered a Bayesian analog of the penalized likelihood approach of Huang et al. (2006) (also Yuan and Lin 2007).”
MethodMinimum Distance Estimation of Milky Way Model Parameters and Related Inference
Banerjee, S., Basu, A., Bhattacharya, S., Bose, S., Chakrabarty, D., & Mukherjee, S. S. (2015) · SIAM/ASA Journal on Uncertainty Quantification · doi:10.1137/130935525 · Cited by 5
“regression models in regression frameworks characterized by the number of covariates far exceeding the number of responses (Yuan and Lin, 2007; Simon et al., 2012), and by developing density based distances within the paradigm of semi-supervised or unsupervised learning (Bijral et al.”
BackgroundKim, W. H., Kim, H. J., Adluru, N., & Singh, V. (2016) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2016.268 · Cited by 5
“Using a (cid:96) 1 penalty on the entries of this matrix, under mild conditions, one can guarantee that the maximum likelihood solution will recover the true model [45, 13].”
BackgroundA two-stage approach of gene network analysis for high-dimensional heterogeneous data
Lee, S., Liang, F., Cai, L., & Xiao, G. (2017) · Biostatistics · doi:10.1093/biostatistics/kxx033 · Cited by 5
Estimating a Covariance Matrix for Market Risk Management and the Case of Credit Default Swaps
Neuberg, R., & Glasserman, P. (2017) · Quantitative Finance · doi:10.1080/14697688.2018.1494850 · Cited by 6
“Yuan and Lin (2007) and Friedman et al. (2008) propose to estimate a sparse graph structure by minimising Lnormal, augmented with a lasso term which penalizes large elements in the estimate ∗−1, and effectively sets some of its elements to zero: min ∗ { ln det( ∗) + tr( emp ∗−1) + μ‖ ∗−1‖1 } .”
Generalized information criterion for model selection in penalized graphical modelsPreprint
Abbruzzo, A., Vujačić, I., Wit, E. C., & Mineo, A. M. (2014) · arXiv (Cornell University) · doi:10.48550/arxiv.1403.1249 · Cited by 5
“In particular, we evaluate the performance of the proposed estimator which we called GIC (22), two fold cross validation (CV), the Akaike’s information criterion (AIC), the Schwarz’s information criterion (BIC), the Schwarz’s information criterion adjusted with degrees of freedom in formula (23) (GBIC (24)).”
BackgroundMethodInfluentialGoto, S., & Xu, Y. (2012) · SSRN Electronic Journal · doi:10.2139/ssrn.2024197 · Cited by 5
“8We follow Yuan and Lin (2007) and Rothman, Bickel, Levina, and Zhu (2008) in setting up the likelihood function with the penalty, and employ the “glasso” algorithm of Friedman, Hastie, and Tibshirani (2008) to solve the estimation problem.”
MethodPost-Regularization Inference for Dynamic Nonparanormal Graphical ModelsPreprint
Lu, J., Kolar, M., & Liu, H. (2015) · arXiv (Cornell University) · Cited by 5
Thrampoulidis, C. (2016) · CaltechTHESIS (California Institute of Technology) · doi:10.7907/z998850v · Cited by 17
Bayesian Low Rank and Sparse Covariance Matrix DecompositionPreprint
Zhang, L., Sarkar, A., & Mallick, B. K. (2013) · arXiv (Cornell University) · doi:10.48550/arxiv.1310.4195 · Cited by 5
“In the graphical lasso method (Yuan and Lin 2007; Friedman et al. 2008), anl1 penalty term is assigned toS, which, in a Bayesian framework, is equivalent to a prior distribution as follows S|λ ∝ ∏ j<j′ exp{−λ|Sjj′”
MethodInfluentialFast and Scalable Learning of Sparse Changes in High-Dimensional Gaussian Graphical Model StructurePreprint
Wang, B., Sekhon, A., & Qi, Y. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1710.11223 · Cited by 5
“Sparse Gaussian Graphical Model(sGGM)[15, 18, 32] assumes data samples are independently and identically drawn from Np(μ, ), a multivariate normal distribution with mean μ and covariance matrix .”
BackgroundMethodA Unified Dynamic Approach to Sparse Model SelectionPreprint
Huang, C., & Yao, Y. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1810.03608 · Cited by 5
“Yuan and Lin (2007); Ravikumar et al. (2008) studied (1.1), with the loss function being the negative scaled log-likelihood , and the penalty being the sum of the absolute values of the off-diagonal entries of the precision matrix.”
InfluentialWang, B., Sekhon, A., & Qi, Y. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1806.00548 · Cited by 5
“For homogeneous data samples from a given i -th context, one typical approach is the sparse Gaussian Graphical Model (sGGM) (Lauritzen, 1996; Yuan & Lin, 2007). sGGM assumes samples are independently and identically drawn from N p ( µ ( i ) , Σ ( i ) ) , a multivariate Gaussian distribution with…”
MethodPrecision Matrix Estimation with Noisy and Missing DataPreprint
Fan, R., Jang, B., Sun, Y., & Zhou, S. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1904.03548 · Cited by 5
“…a wide body of work proposing methods to perform precision matrix estimation in the fully observed case, including Meinshausen and Bühlmann (2006), Yuan and Lin (2007), Rothman et al. (2008), Friedman et al. (2008), Banerjee et al. (2008), and Zhou et al. (2010), most of which are essentially a 1…”
MethodFalse Discovery Rates in Biological NetworksPreprint
Yu, L., Kaufmann, T., & Lederer, J. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1907.03808 · Cited by 5
“Besides simplistic correlational approaches, popular estimators are neighborhood selection (Meinshausen and Bühlmann, 2006), which combines node-wise lasso estimates, and graphical lasso (Friedman et al., 2008; Yuan and Lin, 2007), which maximizes an `1penalized log-likelihood.”
MethodJoint Functional Gaussian Graphical ModelsPreprint
Moysidis, I., & Li, B. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2110.06653 · Cited by 5
“Yuan and Lin (2007) propose an `1 penalized loglikelihood method which has the advantage of merging variable selection and precision matrix estimation into one problem.”
MethodSufficient reductions in regression with mixed predictorsPreprint
Bura, E., Forzani, L., Arancibia, R. G., Llop, P., & Tomassi, D. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2110.13091 · Cited by 5
“References for GMMs for low-dimensional mixed data include Lauritzen and Wermuth (1989), Lauritzen (1996), Yuan and Lin (2007), Wainwright and Jordan (2008), and in the high-dimensional setting, Cheng et al. (2014, 2017) and Lee and Hastie (2015).”
MethodEstimation of high-dimensional vector autoregression via sparse precision matrix
Poignard, B., & Asai, M. (2023) · Econometrics Journal · doi:10.1093/ectj/utad003 · Cited by 4
IDGM: an approach to estimate the graphical model of interval-valued data
Wu, Q., Wang, H., & Lu, S. (2024) · Statistics and Computing · doi:10.1007/s11222-024-10504-7 · Cited by 4
“Yuan and Lin (2007) consider the graphical lasso optimization of the precision matrix and infer edges according to the estimated sparse precision matrix.”
Sottile, G., Augugliaro, L., Vinciotti, V., Arancio, W., & Coronnello, C. (2024) · Journal of the Royal Statistical Society Series C (Applied Statistics) · doi:10.1093/jrsssc/qlae052 · Cited by 4
Conditional extremes with graphical models
Farrell, A., Eastoe, E., & Lee, C. M. (2026) · Computational Statistics & Data Analysis · doi:10.1016/j.csda.2026.108452 · Cited by 4
Learning massive-scale partial correlation networks in clinical multiomics studies with HP-ACCORD
Lee, S., Bang, J., Kim, Y., Choi, H., Oh, S., & Won, J. (2025) · The Annals of Applied Statistics · doi:10.1214/25-aoas2089 · Cited by 7
Sparse signal recovery using sparse random projections
Ramchandran, K., & Wang, W. (2009) · Cited by 4
On the existence of the weighted bridge penalized Gaussian likelihood precision matrix estimator
Rothman, A. J., & Forzani, L. (2014) · Electronic Journal of Statistics · doi:10.1214/14-ejs973 · Cited by 4
“A benefit of using q = 1 is that for λ sufficiently large, a subset of the off-diagonal entries in Ω̂λ,1(M) are zero and when S is computed from n independent copies of a p-variate Normal random vector, the locations of non-zero entries in Ω̂λ,1(M) could be used to estimate the edges in a Gaussian graphical model (Yua…”
MethodHigh-frequency financial statistics with parallel R and Intel Xeon Phi coprocessor
Zou, J., & Zhang, H. (2014) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata.2014.7004414 · Cited by 4
“Other related work include [39], [20], and [15].”
BackgroundSu, N., Dai, D., Deng, C., Qian, M., & Deng, M. (2014) · IET Systems Biology · doi:10.1049/iet-syb.2013.0029 · Cited by 4
“It is generally selected by Akaike information criterion (AIC), Bayesian information criterion (BIC) or cross-validation [36, 37].”
BackgroundMethodApproaches to High‐Dimensional Covariance and Precision Matrix Estimations
Fan, J., Liao, Y., & Liu, H. (2016) · Wiley eBooks · doi:10.1002/9781118745540.ch6 · Cited by 4
“A natural approach for estimating Θ is by penalizing the likelihood using the L1-penalty (Banerjee et al. 2008; Friedman et al. 2008; Yuan and Lin 2007).”
MethodSparse covariance estimation in logit mixture models
Aboutaleb, Y. M., Danaf, M., Xie, Y., & Ben-Akiva, M. E. (2021) · Econometrics Journal · doi:10.1093/ectj/utab008 · Cited by 4
“selection and estimation simultaneously. The implementation of the penalty methods, however, is nontrivial because of the positive denite constraint on the covariance (or concentration) matrix; see Yuan and Lin (2007). Yuan and Lin (2007) proposed a penalized likelihood method for model selection and parameter estimat…”
MethodMultiple and multilevel graphical models
Kim, I., Shan, L., Lin, J., Gao, W., Kim, B., & Mahmoud, H. F. F. (2020) · Wiley Interdisciplinary Reviews Computational Statistics · doi:10.1002/wics.1497 · Cited by 4
“Yuan and Lin (2007) used the determinant maximization (MAXDET) algorithm.”
MethodAnalysis of Twitter data with the Bayesian fused graphical lasso
Aflakparast, M., de Gunst, M. C. M., & van Wieringen, W. N. (2020) · PLoS ONE · doi:10.1371/journal.pone.0235596 · Cited by 4
Xiao, Y., Li, P., & Lu, S. (2021) · Journal of the Operations Research Society of China · doi:10.1007/s40305-021-00351-y · Cited by 4
Jewel: A Novel Method for Joint Estimation of Gaussian Graphical Models
Angelini, C., Canditiis, D. D., & Plaksienko, A. (2021) · Mathematics · doi:10.3390/math9172105 · Cited by 4
Cao, X., & Lee, K. (2023) · Bayesian Analysis · doi:10.1214/23-ba1376 · Cited by 5
“In the frequentist setting, one of the most popular methods to achieve a sparse estimate of Ω is the graphical lasso [Friedman et al., 2007, Yuan and Lin, 2007], where the objective function is composed of the negative Gaussian log-likelihood and an `1-penalty term for the off-diagonal entries of…”
MethodAn Additive Graphical Model for Discrete Data
Tao, J., Li, B., & Xue, L. (2022) · Journal of the American Statistical Association · doi:10.1080/01621459.2022.2119983 · Cited by 4
“…(copula) graphical model setting, a well-known method for estimating a positive definite sparse precision matrix is via the penalized Gaussian likelihood estimator or its rank-based alternative (Yuan and Lin 2007; Friedman, Hastie, and Tibshirani 2008; Fan, Feng, and Wu 2009; Xue and Zou 2012).”
MethodPrecision matrix estimation using penalized Generalized Sylvester matrix equation
Avagyan, V. (2022) · Test · doi:10.1007/s11749-022-00807-0 · Cited by 4
“For this study, we select the penalty parameters for the considered methods using Bayesian Information Criterion (BIC) (Yuan and Lin 2007).”
BackgroundMethodInfluentialLatent Multimodal Functional Graphical Model Estimation
Tsai, K. H., Zhao, B., Koyejo, S., & Kolar, M. (2023) · Journal of the American Statistical Association · doi:10.1080/01621459.2023.2252142 · Cited by 10
“Thus, many high-dimensional methods, such as graphical lasso (Yuan and Lin 2007), are not easily applied, as they require restrictions on correlations to achieve the desired statistical properties.”
Estimation of large covariance matrices with mixed factor structures
Dai, R., Uematsu, Y., & Matsuda, Y. (2023) · Econometrics Journal · doi:10.1093/ectj/utad018 · Cited by 4
Goodness-of-fit tests for high-dimensional Gaussian graphical models via exchangeable sampling
Lin, X., Li, W., Tian, F., & Huang, D. (2025) · Journal of the Royal Statistical Society Series B (Statistical Methodology) · doi:10.1093/jrsssb/qkaf032 · Cited by 4
Learning Networks From Wide-Sense Stationary Stochastic Processes
Rayas, A., Cheng, J., Anguluri, R., Deka, D., & Dasarathy, G. (2025) · IEEE Transactions on Signal and Information Processing over Networks · doi:10.1109/tsipn.2025.3583488 · Cited by 4
“Pioneering works like [30], [31] have developed key theoretical concepts for analyzing (cid:4) 1 -regularized likelihood estimators, and our analysis builds on these concepts.”
The functional graphical lasso
Waghmare, K. G., Masak, T., & Panaretos, V. M. (2025) · The Annals of Statistics · doi:10.1214/25-aos2521 · Cited by 4
“In practice, the graphical lasso is arguably the method of choice for Gaussian graphical models, likely due to its conceptual simplicity and ability to perform estimation and model selection (i.e. support estimation) in a single step (Yuan and Lin, 2007).”
Exact Hybrid Covariance Thresholding for Joint Graphical Lasso
Tang, Q., Yang, C., Peng, J., & Xu, J. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-23525-7_36 · Cited by 4
“This problem can be formulated as graphical lasso [1, 22] and many algorithms [4, 16, 9, 19, 18] have been proposed to solve it.”
BackgroundMarjanovic, G., & Solo, V. (2011) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2011.5946736 · Cited by 5
“Index Terms— Sparse, inverse covariance (precision) matrix, l0 penalty, l1 penalty, loglikelihood function, nonconvex optimization.”
BackgroundModularized Gaussian Graphical Model
Guo, J. F., & Wang, S. (2010) · Cited by 4
Sparse Markov Net Learning with Priors on Regularization Parameters
Scheinberg, K., Rish, I., & Asadi, N. B. (2010) · ISAIM · Cited by 4
Joint estimation of multiple high‐dimensional Gaussian copula graphical models
He, Y., Zhang, X., Ji, J., & Liu, B. (2017) · Australian & New Zealand Journal of Statistics · doi:10.1111/anzs.12198 · Cited by 4
“Yuan & Lin (2007) proposed a penalised likelihood method which is based using on interior point optimisation to directly estimate .”
MethodNovel model selection criteria on sparse biological networks
Bülbül, G. B., Purutçuoğlu, V., & Purutçuoğlu, E. (2019) · International Journal of Environmental Science and Technology · doi:10.1007/s13762-019-02206-9 · Cited by 4
Differential Network Analysis via Weighted Fused Conditional Gaussian Graphical Model
Ou-Yang, L., Zhang, X.-F., Hu, X. T., & Yan, H. (2019) · IEEE Transactions on Computational Biology and Bioinformatics · doi:10.1109/tcbb.2019.2924418 · Cited by 4
“Two variables are conditionally independent (independent given the remaining variables) if and only if the corresponding element of the precision matrix is zero [4].”
BackgroundClustered Gaussian Graphical Model Via Symmetric Convex Clustering
Yao, T., & Allen, G. I. (2019) · IEEE Data Science Workshop (DSW) · doi:10.1109/dsw.2019.8755774 · Cited by 4
Selection by partitioning the solution paths
Liu, Y., & Wang, P. (2018) · Electronic Journal of Statistics · doi:10.1214/18-ejs1434 · Cited by 4
“Criterion Method References CV Adaptive lasso Zou (2006), JASA Fused lasso Tibshirani et al. (2005), JRSSB Truncated l 1 penalty Shen, Pan and Zhu (2012), JASA GCV Lasso Tibshirani (1996), JRSSB SCAD Fan and Li (2001), JASA BIC Lasso Wang, Li and Tsai (2007), JRSSB Lasso Yuan and Lin (2007), Biometrika EBIC Tilting Ch…”
BackgroundMethodInfluentialStructure Learning with Side Information: Sample Complexity
Sihag, S., & Tajer, A. (2019) · neural information processing systems · Cited by 4
High-dimensional generalized linear models incorporating graphical structure among predictors
Zhou, S., Zhou, J., & Zhang, B. (2019) · Electronic Journal of Statistics · doi:10.1214/19-ejs1601 · Cited by 4
“If the prior information cannot be obtained in some applications, we can construct the predictor graph by sparse estimation of the covariance (or precision) matrix of the predictors [39, 11, 6].”
BackgroundMethodA Statistically Efficient and Scalable Method for Exploratory Analysis of High-Dimensional Data
Rahman, M. S., & Haffari, G. (2020) · SN Computer Science · doi:10.1007/s42979-020-0064-2 · Cited by 4
A two-step method for estimating high-dimensional Gaussian graphical models
Yang, Y., & Zhu, J. (2020) · Science China Mathematics · doi:10.1007/s11425-017-9438-5 · Cited by 6
A Novel Network Modelling for Metabolite Set Analysis: A Case Study on CRC Metabolomics
Liu, Y., Xu, X., Deng, L., Cheng, K., Xu, J., Raftery, D., & Dong, J. (2020) · IEEE Access · doi:10.1109/access.2020.3000432 · Cited by 4
Correlation Imputation in Single cell RNA-seq using Auxiliary Information and Ensemble Learning
Gan, L., Vinci, G., & Allen, G. I. (2020) · ACM International Conference on Bioinformatics, Computational Biology, and Health Informatics (ACM-BCB) · doi:10.1145/3388440.3412462 · Cited by 5
“Finally, in Figure 4C we display the conditional dependence graph (graphical lasso; [36]) of the 30 most variable genes among cell types (ANOVA criterion as in Section 3.”
MethodCollege Students’ Psychological Health Analysis Based on Multitask Gaussian Graphical Models
Tian, Q., Wang, R., Li, S., Wang, W., Wu, O., Li, F., & Jiao, P. (2021) · Complexity · doi:10.1155/2021/5710459 · Cited by 4
Detecting clusters in multivariate response regression
Price, B. S., Allenbrand, C., & Sherwood, B. (2021) · Wiley Interdisciplinary Reviews Computational Statistics · doi:10.1002/wics.1551 · Cited by 4
“…graphical models, and the representation of the inverse covariance matrix as a graphical model, we refer the reader to d'Aspermont et al. (2008), Yuan and Lin (2007), Friedman et al. (2007), Witten et al. (2011), Loh and Wainwright (2012) and others who have grown the literature in this area.”
MethodContraction of a quasi-Bayesian model with shrinkage priors in precision matrix estimation
Zhang, R., Yao, Y., & Ghosh, M. (2022) · Journal of Statistical Planning and Inference · doi:10.1016/j.jspi.2022.03.003 · Cited by 4
“Penalized likelihood estimators are proposed in Huang et al. (2006), Yuan and Lin (2007), Banerjee et al. (2008), Friedman et al. (2008), Zhang and Zou (2014), among others.”
BackgroundMethodA Cholesky-based sparse covariance estimation with an application to genes data
Li, C., Yang, M., Wang, M., Kang, H., & Kang, X. (2021) · Journal of Biopharmaceutical Statistics · doi:10.1080/10543406.2021.1931270 · Cited by 4
“Therefore, we obtain a sparse estimate ( Then a sparse estimate for the covariance matrix is ~� τ The tuning parameter τ can be selected by minimizing the BIC (Yuan and Lin 2007) where e ij ð τ Þ ¼ 0 if ð ~ σ τ Þ ij ¼ 0, and e ij ð τ Þ ¼ 1 otherwise.”
MethodResultInfluentialOn graphical models and convex geometry
Bar, H., & Wells, M. T. (2023) · Computational Statistics & Data Analysis · doi:10.1016/j.csda.2023.107800 · Cited by 4
“See, for example Meinshausen and B¨uhlmann (2006); Friedman et al. (2008); Yuan and Lin (2007); Peng et al. (2009); Khare et al. (2015).”
BackgroundA new double-regularized regression using Liu and lasso regularization
Genç, M. (2021) · Computational Statistics · doi:10.1007/s00180-021-01120-4 · Cited by 4
“Some generalizations are grouped lasso (Yuan and Lin 2006), graphical lasso (Yuan and Lin 2007), nearly isotonic regression (Tibshirani et al. 2011), near-optimal matrix completion (Candès and Tao 2010) and square-root lasso (Belloni et al. 2014) among others.”
MethodDegano, I., & Lotito, P. A. (2021) · Transactions in GIS · doi:10.1111/tgis.12799 · Cited by 5
Confidence graphs for graphical model selection
Wang, L., Qin, Y., & Li, Y. (2021) · Statistics and Computing · doi:10.1007/s11222-021-10027-5 · Cited by 4
A generative approach to modeling data with quantitative and qualitative responses
Kang, X., Kang, L., Chen, W., & Deng, X. (2022) · Journal of Multivariate Analysis · doi:10.1016/j.jmva.2022.104952 · Cited by 4
“Next, we show that (7) can be decomposed as a graphical lasso model (Glasso) (Yuan and Lin, 2007; Deng and Yuan, 2009) in terms of C and a Lasso regression (Tibshirani, 1996) in terms of δ2 with the other parameter fixed, such that these two parameters can be estimated iteratively. To be more precise, for a given valu…”
InfluentialNetDA: An R Package for Network-Based Discriminant Analysis Subject to Multilabel Classes
Chen, L. (2022) · Journal of Probability and Statistics · doi:10.1155/2022/1041752 · Cited by 4
“By the similar discussion in Yuan and Lin [14], the estimated network structure determined by (5) is equal to the true graph with probability approaching one under suitable conditions.”
BackgroundZhao, H., Datta, S., & Duan, Z.-H. (2023) · Bioinformatics and Biology Insights · doi:10.1177/11779322231152972 · Cited by 5
Extending compositional data analysis from a graph signal processing perspective
Rieser, C., & Peter Filzmoser, A. (2023) · Journal of Multivariate Analysis · doi:10.1016/j.jmva.2023.105209 · Cited by 4
Bayesian Learning of Relational Graphs in Semiparametric High-dimensional Time Series
Roy, A., Roy, A., & Ghosal, S. (2026) · Bayesian Analysis · doi:10.1214/26-ba1589 · Cited by 4
Topology Adaptive Graph Estimation in High Dimensions
Lederer, J., & Müller, C. L. (2022) · Mathematics · doi:10.3390/math10081244 · Cited by 4
“In practice, the tuning parameters are calibrated via cross-validation, classical information criteria such as the AIC and BIC [8], or stability criteria [18].”
BackgroundMethodInfluentialRecent developments of constructing adjacency matrix in network analysis
Hong, Y., & Kim, C. (2014) · Journal of the Korean Data and Information Science Society · doi:10.7465/jkdi.2014.25.5.1107 · Cited by 6
“Yuan and Lin (2007) discussed penalized maximum likelihood with a lasso penalty on inverse of covariance matrix.”
BackgroundContrastive Structured Anomaly Detection for Gaussian Graphical Models
Maurya, A., & Cheung, M. (2018) · IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM) · doi:10.1109/asonam.2018.8508475 · Cited by 4
“This estimation can be performed using several methods such as selection procedure (Scheinberg et al., 2010) and graphical lasso (Yuan & Lin, 2007).”
MethodInfluentialMultivariate semiparametric spatial methods for imaging data
Chen, H., Cao, G., & Cohen, R. A. (2016) · Biostatistics · doi:10.1093/biostatistics/kxw052 · Cited by 4
“Interestingly, the inverse covariance matrix −1 is the concentration matrix, which measures the partial correlation between two ROIs conditional all the other R − 2 ROIs (Yuan and Lin, 2007).”
BackgroundMethodFused lasso regression for identifying differential correlations in brain connectome graphs
Yu, D., Lee, S. H., Lim, J., Xiao, G., Craddock, R. C., & Biswal, B. B. (2018) · Statistical Analysis and Data Mining The ASA Data Science Journal · doi:10.1002/sam.11382 · Cited by 4
A structural model on a hypercube represented by optimal transport
Sei, T. (2011) · Statistica Sinica · doi:10.5705/ss.2009.022 · Cited by 4
“The graphical Gaussian lasso estimator Ĉ = Ĉ(τ) of the concentration matrix (Yuan and Lin (2007)) is formulated as”
BackgroundMethodInfluentialAsymptotic results for multivariate local Whittle estimation with applications
Düker, M., & Pipiras, V. (2019) · IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing · doi:10.1109/camsap45676.2019.9022642 · Cited by 4
Chen, X. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1602.00199 · Cited by 4
“Recently, due to the explosive data enrichment, regularized estimation and dimension reduction of high-dimensional data have attracted a lot of research attentions such as covariance matrix estimation [6, 7, 22, 14], graphical models [20, 56, 9], discriminant analysis [38], factor models [23, 34], among many others.”
BackgroundLocal Optima of Nonconvex Regularized M-Estimators
Loh, P. (2013) · UC Berkeley · Cited by 4
A partial orthogonalization method for simulating covariance and concentration graph matricesPreprint
Córdoba, I., Varando, G., Bielza, C., & Larrañaga, P. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1807.03090 · Cited by 4
“…at the literature on these models, one finds that typical benchmark structures are Toeplitz, banded, diagonally spiked and block diagonal covariance or concentration matrices (Yuan and Lin, 2007; Xue and Zou, 2012; Ledoit and Wolf, 2012), with parameters fixed to ensure positive definiteness.”
BackgroundEfficient Covariance Estimation from Temporal DataPreprint
Harutyunyan, H., Moyer, D., Khachatrian, H., Steeg, G. V., & Galstyan, A. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1905.13276 · Cited by 4
“The problem of network inference using graphical lasso is well studied [13, 23, 28, 35].”
MethodEstimation of sparse Gaussian graphical models with hidden clustering structurePreprint
Lin, M., Sun, D., Toh, K.-C., & Wang, C. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2004.08115 · Cited by 4
“To detect nonzero elements in the concentration matrix Σ−1, researchers have proposed sparse Gaussian graphical models [1, 33].”
BackgroundMethodInfluentialRegularized joint mixture modelsPreprint
Perrakis, K., Lartigue, T., Dondelinger, F., & Mukherjee, S. (2019) · Lancaster EPrints (Lancaster University) · doi:10.48550/arxiv.1908.07869 · Cited by 4
“1 Regularization of Σk For the regularization of Σk we use the graphical lasso (Meinshausen and Bühlmann, 2006; Yuan and Lin, 2007; Friedman et al., 2008a).”
MethodBayesian Spanning Tree: Estimating the Backbone of the Dependence GraphPreprint
Duan, L. L., & Dunson, D. B. (2021) · arXiv (Cornell University) · doi:10.13140/rg.2.2.14469.88808 · Cited by 4
Phan, D. N. (2016) · theses.fr (ABES) · doi:10.70675/4449a8b5z5dc4z453bzb0b0zc4485be866f4 · Cited by 5
Wang, T., Guo, A., Shima, K., & Ohno, Y. (2025) · Frontiers in Bioengineering and Biotechnology · doi:10.3389/fbioe.2025.1543468 · Cited by 3
High-Frequency Financial Statistics Through High-Performance Computing
Zou, J., & Zhang, H. (2016) · Springer eBooks · doi:10.1007/978-3-319-33742-5_11 · Cited by 3
Dong, X., Thanou, D., Rabbat, M., & Frossard, P. (2021) · Wiley eBooks · doi:10.1002/9781119850830.ch2 · Cited by 3
Pritikin, J. N., & Schmidt, K. M. (2022) · International Journal of Applied Positive Psychology · doi:10.1007/s41042-022-00071-5 · Cited by 3
“Instead of relying on a statistical procedure for item selection (e.g., Tibshirani 1996; Yuan & Lin 2007), we selected items by inspection of the Gaussian graphical model (Epskamp et al.”
MethodDifferential Analysis for Networks Obeying Conservation Laws
Rayas, A., Anguluri, R., Cheng, J., & Dasarathy, G. (2023) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49357.2023.10094876 · Cited by 3
“Θ ∗ i ) 12 using well-known estimators such as GLASSO or CLIME [13, 14, 15].”
MethodLin, J., & Kim, I. (2024) · Statistical Analysis and Data Mining The ASA Data Science Journal · doi:10.1002/sam.11699 · Cited by 3
“We also compare the performance between Group Lasso [26, 48] and our GPS.”
ResultClusterpath Gaussian Graphical Modeling
Touw, D. J. W., Alfons, A., Groenen, P. J. F., & Wilms, I. (2026) · Journal of Computational and Graphical Statistics · doi:10.1080/10618600.2026.2653764 · Cited by 3
Sparse Inverse Covariance Estimation for Graph Representation of Feature Structure
Lee, S. (2014) · Lecture notes in computer science · doi:10.1007/978-3-662-43968-5_13 · Cited by 3
Effectiveness of combinations of Gaussian graphical models for model building
Massa, M. S., & Chiogna, M. (2012) · Journal of Statistical Computation and Simulation · doi:10.1080/00949655.2012.667216 · Cited by 3
Analysis for temporal gene expressions under multiple biological conditions
Fang, H., Deng, D., Tian, G., Shen, L., Duan, K., & Song, J. (2012) · Statistics in Biosciences · doi:10.1007/s12561-012-9063-8 · Cited by 7
“The high-throughput gene expression techniques, such as oligonucleotide and DNA microarray, serial analysis of gene expression, etc., make it possible now to quickly generate huge amounts of time series data on gene expression under various conditions [2, 3, 16, 28, 31].”
BackgroundToward Breaking the Histone Code: Bayesian Graphical Models for Histone Modifications
Mitra, R., Müller, P., Liang, S., Xu, Y., & Ji, Y. (2013) · Circulation Cardiovascular Genetics · doi:10.1161/circgenetics.113.000100 · Cited by 4
Node-Structured Integrative Gaussian Graphical Model Guided by Pathway Information
Kim, S., Jhong, J., Lee, J., Koo, J., Lee, B., & Han, S. (2017) · Computational and Mathematical Methods in Medicine · doi:10.1155/2017/8520480 · Cited by 3
“The optimal value for λ can be chosen by cross-validation or the Bayesian information criterion (BIC; Schwarz [18]; Yuan and Lin [8]).”
Method
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.