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 501–600 of 1,000
A proximal distance algorithm for likelihood-based sparse covariance estimation
Xu, J., & Lange, K. (2022) · Biometrika · doi:10.1093/biomet/asac011 · Cited by 23
“Penalized likelihood techniques offer an alternative to thresholding and are arguably the preferred method for estimating sparse precision (inverse covariance) matrices [Yuan and Lin, 2007, Molstad and Rothman, 2018].”
MethodVariable Selection for Gaussian Graphical Models
Honorio, J., Samaras, D., Rish, I., & Cecchi, G. (2012) · International Conference on Artificial Intelligence and Statistics · Cited by 17
Learning a factor model via regularized PCA
Kao, Y.-H., & Roy, B. V. (2013) · Machine Learning · doi:10.1007/s10994-013-5345-8 · Cited by 18
“Similar formulations can also be found in Yuan and Lin (2007) and Ravikumar et al. (2011), who instead penalize the ℓ1 norm of off-diagonal elements of the inverse covariance matrix when computing maximum-likelihood estimates.”
BackgroundEfficient Bayesian Regularization for Graphical Model Selection
Kundu, S., Mallick, B. K., & Baladandayuthapani, V. (2018) · Bayesian Analysis · doi:10.1214/17-ba1086 · Cited by 17
“For our approach, the point estimate is obtained under a false discovery rate of 0.2, whereas the estimates for the frequentist approaches were obtained by minimizing a Bayesian Information Criteria or BIC, as in Yuan and Lin (2007).”
MethodInfluentialA GICA-TVGL framework to study sex differences in resting state fMRI dynamic connectivity
Cai, B., Zhang, G., Zhang, A., Hu, W., Stephen, J. M., Wilson, T. W., Calhoun, V. D., & Wang, Y. (2019) · Journal of Neuroscience Methods · doi:10.1016/j.jneumeth.2019.108531 · Cited by 17
Robust Gaussian Graphical Modeling with the Trimmed Graphical LassoPreprint
Yang, E., & Lozano, A. C. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1510.08512 · Cited by 17
“[9]) minimizes the Gaussian negative log-likelihood regularized by the `1 norm of the entries (or the off-diagonal entries) of the precision matrix (see [10, 11, 12]).”
BackgroundNonparametric Functional Graphical Modeling Through Functional Additive Regression Operator
Lee, K., Li, L., Li, B., & Zhao, H. (2021) · Journal of the American Statistical Association · doi:10.1080/01621459.2021.2006667 · Cited by 16
He, Y., Li, Q., Hu, Q., & Liu, L. (2022) · Statistics in Medicine · doi:10.1002/sim.9499 · Cited by 24
“…penalized methods, see for example, neighborhood selection method by Meinshausen and B¨uhlmann (2006), graphical Lasso by Friedman et al. (2008); Yuan and Lin (2007), the lasso penalized D-trace loss by Zhang and Zou (2014), and constrained (cid:96) 1 minimization approach (CLIME) by Cai et al.…”
MethodHou, D., Zhou, W., Zhang, Q., Zhang, K., & Fang, J. (2023) · PeerJ Computer Science · doi:10.7717/peerj-cs.1522 · Cited by 16
Estimation of Graphical Models: An Overview of Selected Topics
Chen, L. (2023) · International Statistical Review · doi:10.1111/insr.12552 · Cited by 18
Gao, X., & Massam, H. (2014) · Journal of Computational and Graphical Statistics · doi:10.1080/10618600.2014.937811 · Cited by 16
“Yuan and Lin (2007) proposed penalized likelihood methods for estimating the concentration matrix with the L 1 LASSO penalty.”
MethodCopula Gaussian graphical models with penalized ascent Monte Carlo EM algorithm
Abegaz, F., & Wit, E. C. (2015) · Statistica Neerlandica · doi:10.1111/stan.12066 · Cited by 16
“More-over, we extend this approach to a high-dimensional inference based on graphical models using 𝓁 l -penalized maximum likelihood ( Banerjee et al. , 2008; Friedman et al. , 2008; Rothman et al. , 2008; Yuan and Lin , 2007).”
MethodOn an additive partial correlation operator and nonparametric estimation of graphical models
Lee, K., Li, B., & Zhao, H. (2016) · Biometrika · doi:10.1093/biomet/asw028 · Cited by 16
“…under the receiver operating characteristic curves for the additive partial correlation operator have an average of 0 · 98 for the two curves in Model I, only slightly smaller than the average of the areas under the curves for the methods of Yuan & Lin (2007) and Liu et al. (2009), which is 1 · 00.”
BackgroundMethodInfluentialScene Labeling Using Sparse Precision Matrix
Souly, N., & Shah, M. A. (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.2016.397 · Cited by 16
“To achieve sparsity, [30] proposed to use a lasso (Least Absolute Shrinkage and Selection Operator) model [24] to estimate each variables using others as predictor and by applying L1 regularization on coefficients to enforce sparsity.”
MethodSparse precision matrices for minimum variance portfolios
Torri, G., Giacometti, R., & Paterlini, S. (2019) · Computational Management Science · doi:10.1007/s10287-019-00344-6 · Cited by 16
“…a square n ×n matrix with the off-diagonal elements equal to the corresponding elements of Ω and the diagonal elements equal to zero (Friedman et al. 2008).1 As pointed out by Yuan and Lin (2007), the use of an L1 penalty allows to induce sparsity in the precision matrix, making it possible to…”
Graph-based sparse linear discriminant analysis for high-dimensional classification
Liu, J., Yu, G., & Liu, Y. (2018) · Journal of Multivariate Analysis · doi:10.1016/j.jmva.2018.12.007 · Cited by 17
“With recent advances in graphical estimation [7, 28, 46], we can estimate 𝒢 for the GSLDA based on the training data when the graph structure is unknown.”
BackgroundMethodInfluentialSecond-Order Conditional Gradient Sliding
Carderera, A., & Pokutta, S. (2026) · Fields Institute communications · doi:10.1007/978-3-032-03844-9_12 · Cited by 16
Brain connectivity alteration detection via matrix‐variate differential network model
Ji, J., He, Y., Liu, L., & Xie, L. (2020) · Biometrics · doi:10.1111/biom.13359 · Cited by 17
Functional directed graphical models and applications in root-cause analysis and diagnosis
Gómez, A. M. E., Paynabar, K., & Pacella, M. (2020) · Journal of Quality Technology · doi:10.1080/00224065.2020.1805380 · Cited by 16
Nonparametric and high-dimensional functional graphical models
Solea, E., & Dette, H. (2022) · Electronic Journal of Statistics · doi:10.1214/22-ejs2087 · Cited by 17
Lederer, J. (2021) · Springer texts in statistics · doi:10.1007/978-3-030-73792-4_2 · Cited by 18
Regularized parameter estimation of high dimensional t distribution
Yuan, M., & Huang, J. Z. (2008) · Journal of Statistical Planning and Inference · doi:10.1016/j.jspi.2008.10.014 · Cited by 16
“A number of approaches (Ledoit and Wolf, 2004; Huang et al., 2006; Yuan and Lin, 2007; Bickel and Levina, 2008) have been introduced in recent years to overcome the difficulties associated with high dimensions.”
Self-citationGINI: From ISH Images to Gene Interaction Networks
Puniyani, K., & Xing, E. P. (2013) · PLoS Computational Biology · doi:10.1371/journal.pcbi.1003227 · Cited by 16
“For p~50 dimensions, the true inverse covariance matrix was constructed by using the AR(1) model from [42].”
MethodOn ${l}_{q}$ Optimization and Sparse Inverse Covariance Selection
Marjanovic, G., & Solo, V. (2014) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2014.2303429 · Cited by 16
“The MCC is considered in [11], [22], [33], and is defined as:”
BackgroundMethodInfluentialAn algorithm for the multivariate group lasso with covariance estimation
Wilms, I., & Croux, C. (2017) · Journal of Applied Statistics · doi:10.1080/02664763.2017.1289503 · Cited by 16
“We use the Bayesian Information Criterion to select the optimal value of the sparsity parameter λ ω (e.g. Yuan and Lin, 2007).”
Sparse Multi-Dimensional Graphical Models: A Unified Bayesian Framework
Ni, Y., Stingo, F. C., & Baladandayuthapani, V. (2017) · Journal of the American Statistical Association · doi:10.1080/01621459.2016.1167694 · Cited by 20
“…Sho-jaie and Michailidis 2010; Stingo et al. 2010; Yajima et al. 2015) and undirected graphs (Dobra et al. 2004; Meinshausen and Bühlmann 2006; Yuan and Lin 2007; Banerjee, El Ghaoui, and d’Aspremont 2008; Friedman, Hastie, and Tibshirani 2008; Car-valho and Scott 2009; Kundu,…”
BackgroundPrecision Matrix Estimation in High Dimensional Gaussian Graphical Models with Faster Rates
Wang, L., Ren, X., & Gu, Q. (2016) · International Conference on Artificial Intelligence and Statistics · Cited by 32
Sparse Inverse Covariance Estimation with L 0 Penalty for Network Construction with Omics Data
Liu, Z., Lin, S., Deng, N., McGovern, D. P. B., & Piantadosi, S. (2016) · Journal of Computational Biology · doi:10.1089/cmb.2015.0102 · Cited by 16
“…because of the computationally challenging and lack of efficient method for L 0 optimization, the popular approach is to optimize the convex relaxation j O j 1 , where j F j 1 indicates the element-wise L 1 norm of the matrix (Banerjee et al., 2008; Friedman et al., 2008; Yuan and Lin, 2007).”
$\ell_1$ -Penalized censored Gaussian graphical model
Augugliaro, L., Abbruzzo, A., & Vinciotti, V. (2018) · Biostatistics · doi:10.1093/biostatistics/kxy043 · Cited by 22
“The graphical lasso (Yuan and Lin, 2007) is a popular method for estimating a sparse concentration matrix, based on the idea of adding”
MethodNonstationary Modeling With Sparsity for Spatial Data via the Basis Graphical Lasso
Krock, M., Kleiber, W., & Becker, S. (2020) · Journal of Computational and Graphical Statistics · doi:10.1080/10618600.2020.1811103 · Cited by 17
“Yuan and Lin (2007) and Banerjee, El Ghaoui, and d’Aspremont (2008) both use interior point methods, but the latter examine the dual problem of (13) where | U | ≤ (cid:7) is understood elementwise, and solve (14) one column at a time via quadratic programming.”
BackgroundInformation‐incorporated Gaussian graphical model for gene expression data
Yi, H., Zhang, Q., Lin, C., & Ma, S. (2021) · Biometrics · doi:10.1111/biom.13428 · Cited by 16
Joint estimation of multiple Gaussian graphical models across unbalanced classes
Shan, L., & Kim, I. (2017) · Computational Statistics & Data Analysis · doi:10.1016/j.csda.2017.11.009 · Cited by 16
Shah, R. D., Frot, B., Thanei, G.-A., & Meinshausen, N. (2020) · Journal of the Royal Statistical Society Series B (Statistical Methodology) · doi:10.1111/rssb.12359 · Cited by 22
“One approach to sparse precision matrix estimation in the absence of confounding is the graphical lasso (Yuan and Lin, 2007; Yuan, 2010; Friedman et al. , 2008).”
BackgroundDISCUSSION: LATENT VARIABLE GRAPHICAL MODEL SELECTION VIA CONVEX OPTIMIZATION1Preprint
Yuan, M. (2012) · Cited by 16
Self-citationLearning non-Gaussian graphical models via Hessian scores and triangular transportPreprint
Baptista, R., Marzouk, Y., Morrison, R., & Zahm, O. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2101.03093 · Cited by 19
“One of the best-known methods to tackle this problem is the graphical lasso ( glasso ), introduced by Banerjee et al. (2008); Yuan and Lin (2007), which solves an (cid:96) 1 -penalized maximum likelihood estimation problem for the precision matrix.”
MethodNetwork Filtering of Spatial-temporal GNN for Multivariate Time-series Prediction
Wang, Y., & Aste, T. (2022) · ACM International Conference on AI in Finance (ICAIF) · doi:10.1145/3533271.3561678 · Cited by 15
“Similar methods that maximizes 𝐿 1 penalized log-likelihood have been studies by Yuan and Lin [65] and Banerjee et al. [3].”
MethodIdentifying the determinants of lapse rates in life insurance: an automated Lasso approach
Reck, L., Schupp, J., & Reuß, A. (2022) · European Actuarial Journal · doi:10.1007/s13385-022-00325-1 · Cited by 15
“Yuan and Lin [37] apply a grouped Lasso, where all or no coefficient of a covariate remain in the model.”
BackgroundMethodInferring networks from time series: A neural approach
Gaskin, T., Pavliotis, G. A., & Girolami, M. (2024) · PNAS Nexus · doi:10.1093/pnasnexus/pgae063 · Cited by 19
“In an entirely different setting, networks crop up in statistics in the form of conditional independence graphs, describing dependencies between different variables, which again are to be inferred from data (24, 25).”
Data Science Methods for Real-World Evidence Generation in Real-World Data
Liu, F. (2024) · Annual Review of Biomedical Data Science · doi:10.1146/annurev-biodatasci-102423-113220 · Cited by 17
Bayesian sparse graphical models and their mixtures
Talluri, R., Baladandayuthapani, V., & Mallick, B. K. (2014) · Stat · doi:10.1002/sta4.49 · Cited by 15
“…A D 1 n (i.e. a matrix of all 1’s), this gives rise to the special case of the Bayesian version of the graphical lasso of Friedman et al. (2008) and Yuan & Lin (2007), where the single penalty parameter ( (cid:7) ) controls the sparsity of the graph and is estimated via cross-validation or by…”
BackgroundMethodFactorial graphical models for dynamic networks
Wit, E. C., & Abbruzzo, A. (2015) · Network Science · doi:10.1017/nws.2015.2 · Cited by 17
“Various penalized maximum likelihood approaches using the (cid:2) 1 - penalty have been considered (Banerjee et al., 2008; Friedman et al., 2008; Rothman et al., 2008; Yuan & Lin, 2007).”
MethodIdentifying differential networks based on multi-platform gene expression data
Ou-Yang, L., Yan, H., & Zhang, X.-F. (2016) · Molecular BioSystems · doi:10.1039/c6mb00619a · Cited by 15
Bayesian Inference in Nonparanormal Graphical Models
Mulgrave, J. J., & Ghosal, S. (2019) · Bayesian Analysis · doi:10.1214/19-ba1159 · Cited by 21
“Many algorithms have been proposed to solve this problem (Meinshausen and Buhlmann, 2006; Yuan and Lin, 2007; Friedman et al., 2008; Banerjee et al., 2008; d’Aspremont et al., 2008; Rothman et al., 2008; Lu, 2009; Scheinberg et al., 2010; Witten et al., 2011; Mazumder and Hastie, 2012).”
MethodGraph Estimation for Matrix-variate Gaussian Data
Chen, X., & Liu, W. (2017) · Statistica Sinica · doi:10.5705/ss.202017.0076 · Cited by 21
“vectorvariate GGMs. The vector-variate GGM with random vector observations can be viewed as a special case of MGGM with p= 1 or q= 1 and readers may refer to, e.g., Meinshausen and Buhlmann (2006); Yuan and Lin (2007); Rothman et al. (2008); d’Aspremont et al. (2008); Friedman et al. (2008); Yuan (2010); Ravikumar et…”
BackgroundCertifiably optimal sparse inverse covariance estimation
Bertsimas, D., Lamperski, J., & Pauphilet, J. (2019) · Mathematical Programming · doi:10.1007/s10107-019-01419-7 · Cited by 21
Bayesian Joint Estimation of Multiple Graphical Models
Gan, L., Yang, X., Narisetty, N. N., & Liang, F. (2019) · neural information processing systems · Cited by 15
A Selective Overview of Recent Advances in Spectral Clustering and Their Applications
Yang, X., Srinivasan, A., & Xue, L. (2021) · Emerging topics in statistics and biostatistics · doi:10.1007/978-3-030-72437-5_12 · Cited by 17
Discussion: Latent variable graphical model selection via convex optimization
Giraud, C., & Tsybakov, A. B. (2012) · The Annals of Statistics · doi:10.1214/12-aos984 · Cited by 15
Empirical comparison study of approximate methods for structure selection in binary graphical models
Viallon, V., Banerjee, O., Jougla, É., Rey, G., & Coste, J. (2013) · Biometrical Journal · doi:10.1002/bimj.201200253 · Cited by 15
“The two most popular procedures for selecting tuning parameters are the cross-validation (CV) and the Bayesian Information Criterion (BIC), the latter being generally considered as computationally more efficient (Yuan and Lin, 2007).”
MethodMultivariate gene-set testing based on graphical models
Städler, N., & Mukherjee, S. (2014) · Biostatistics · doi:10.1093/biostatistics/kxu027 · Cited by 19
“The Graphical Lasso (Yuan and Lin, 2007; Friedman and others , 2008) estimates the concentration matrix (cid:2) by optimizing where S is the sample covariance matrix and λ denotes a penalty parameter.”
BackgroundInfluentialDaily questionnaire to assess self-reported well-being during a software development project
Kuutila, M., Mäntylä, M., Claes, M., & Elovainio, M. J. (2018) · doi:10.1145/3194932.3194942 · Cited by 17
“Similarly to penalized regression, e.g. Ridge [10] or Lasso regression [18], penalized models have been proposed for GGMs as well [23].”
MethodThresholded graphical lasso adjusts for latent variables
Wang, M., & Allen, G. I. (2022) · Biometrika · doi:10.1093/biomet/asac060 · Cited by 15
Markov Network Estimation From Multi-attribute Data
Kolar, M., Liu, H., & Xing, E. P. (2018) · KiltHub Repository · doi:10.1184/r1/6475871 · Cited by 15
High dimensional Sparse Gaussian Graphical Mixture ModelPreprint
Lotsi, A., & Wit, E. C. (2013) · arXiv (Cornell University) · doi:10.48550/arxiv.1308.3381 · Cited by 15
“The penalized likelihoodbased method (Friedman et al., 2008; Yuan and Lin, 2007) is a promising approach to counter the degeneracy of ly(γ) while keeping the parameter space Γ unaltered.”
MethodA Moreau-Yosida approximation scheme for a class of high-dimensional posterior distributionsPreprint
Atchadé, Y. F. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1505.07072 · Cited by 15
Tensors in Modern Statistical Learning
Sun, W. W., Hao, B., & Li, L. (2021) · Wiley eBooks · doi:10.1002/9781118445112.stat08319 · Cited by 23
“The problem reduces to the classical sparse vector graphical model (Yuan and Lin, 2007; Friedman et al., 2008) when D = 1, and the sparse matrix graphical model (Leng and Tang, 2012; Yin and Li, 2012; Tsiligkaridis et al., 2013; Zhou, 2014) when D = 2.”
BackgroundRegularised Manova for high-dimensional data
Ullah, I., & Jones, B. (2015) · Australian & New Zealand Journal of Statistics · doi:10.1111/anzs.12126 · Cited by 14
“This idea was extended by Yuan & Lin (2007) to estimation of the inverse covariance matrix .”
MethodOn the inconsistency of ℓ 1-penalised sparse precision matrix estimation
Heinävaara, O., Leppä-aho, J., Corander, J., & Honkela, A. (2016) · BMC Bioinformatics · doi:10.1186/s12859-016-1309-x · Cited by 14
“There exists a wide variety of methods making use of (cid:2) 1 -regularisation in the setting of Gaussian graphical model structure learning [6–8”
MethodAgoalikum, E., Klugah‐Brown, B., Yang, H., Wang, P., Varshney, S., Niu, B., & Biswal, B. B. (2021) · Frontiers in Human Neuroscience · doi:10.3389/fnhum.2021.697696 · Cited by 14
“Kindly refer to Meinshausen and Bühlmann (2006); Yuan and Lin (2007), and Banerjee et al. (2008) for more information.”
BackgroundA focused information criterion for graphical models
Pircalabelu, E., Claeskens, G., & Waldorp, L. (2012) · Statistics and Computing · doi:10.1007/s11222-014-9504-y · Cited by 16
“…the lasso procedure to multivariate density estimation, by using ideas of penalization directly on the (cid:5) − 1 matrix, and labeled the resulting procedure as graphical lasso (GLasso), concluding as in Yuan and Lin (2007) that the nodewise procedure is just an approximation to the GLasso.”
MethodAutomatic odor prediction for electronic nose
Mirshahi, M., Nia, V. P., & Adjengue, L. (2018) · Journal of Applied Statistics · doi:10.1080/02664763.2018.1441382 · Cited by 14
“Various techniques were suggested for estimating Θ sparsely, such as Meinshausen and Buhlmann (2006); Yuan and Lin (2007); Banerjee et al. (2008).”
MethodZhang, A., Cai, B., Hu, W., Jia, B., Liang, F., Wilson, T. W., Stephen, J. M., Calhoun, V. D., & Wang, Y. (2019) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2019.2926667 · Cited by 14
Markov Neighborhood Regression for High-Dimensional Inference
Liang, F., Xue, J., & Jia, B. (2020) · Journal of the American Statistical Association · doi:10.1080/01621459.2020.1841646 · Cited by 14
“The existing methods include graphical Lasso (Yuan and Lin, 2007; Friedman et al., 2008), nodewise regression (Meinshausen and Bühlmann, 2006), and ψ-learning (Liang et al.”
BackgroundMethodInfluentialZhang, H., Fu, R., Wang, C., Guo, Y., & Yuan, W. (2022) · IEEE Internet of Things Journal · doi:10.1109/jiot.2022.3188312 · Cited by 14
Chowdhury, S., Wang, R., Yu, Q., Huntoon, C. J., Karnitz, L. M., Kaufmann, S. H., Gygi, S. P., Birrer, M. J., Paulovich, A. G., Peng, J., et al. (2022) · BMC Bioinformatics · doi:10.1186/s12859-022-04864-y · Cited by 17
“As many other models for constructing high-dimensional networks, DAGBagM is designed to infer sparse networks [5, 33, 34].”
BackgroundConvex optimization techniques in system identification
Vandenberghe, L. (2012) · IFAC Proceedings Volumes · doi:10.3182/20120711-3-be-2027.00244 · Cited by 14
“See [Meinshausen and Bühlmann, 2006, Banerjee et al., 2008, Ravikumar et al., 2008, Friedman et al., 2008, Lu, 2009, Scheinberg and Rish, 2009, Yuan and Lin, 2007, Duchi et al., 2008, Li and Toh, 2010, Scheinberg and Ma, 2012].”
BackgroundComputational network analysis of the anatomical and genetic organizations in the mouse brain
Ji, S. (2011) · Bioinformatics · doi:10.1093/bioinformatics/btr558 · Cited by 14
Graphical modeling of binary data using the LASSO: a simulation study
Strobl, R., Grill, E., & Mansmann, U. (2012) · BMC Medical Research Methodology · doi:10.1186/1471-2288-12-16 · Cited by 14
“BIC has shown to be superior as suggested by Yuan and Lin and demonstrated superior performance through simulation studies in Gaussian graphical modes [35,36].”
MethodPrediction in abundant high-dimensional linear regression
Cook, R. D., Forzani, L., & Rothman, A. J. (2013) · Electronic Journal of Statistics · doi:10.1214/13-ejs872 · Cited by 14
“A nat-ural estimator that exploits this condition is that obtained by L 1 -penalized likelihood (Yuan and Lin, 2007; Friedman, Hastie and Tibshirani, 2008; Rothman et al., 2008).”
MethodGrid Topology Identification via Distributed Statistical Hypothesis Testing
Bolognani, S. (2017) · Elsevier eBooks · doi:10.1016/b978-0-12-811968-6.00013-9 · Cited by 14
Joint Mean and Covariance Estimation with Unreplicated Matrix-Variate Data
Hornstein, M., Fan, R., Shedden, K., & Zhou, S. (2018) · Journal of the American Statistical Association · doi:10.1080/01621459.2018.1429275 · Cited by 22
“In the setting where the correlations exist along only one axis of the array, researchers have proposed various covariance estimators and studied their theoretical and numerical properties (Banerjee et al., 2008; Fan et al., 2009; Friedman et al., 2008; Lam and Fan, 2009; Meinshausen and Bühlmann, 2006; Peng et al., 2…”
BackgroundShrinking the Covariance Matrix Using Convex Penalties on the Matrix-Log Transformation
Yi, M., & Tyler, D. E. (2020) · Journal of Computational and Graphical Statistics · doi:10.1080/10618600.2020.1814788 · Cited by 14
“−1}ij| (Yuan and Lin 2007; Friedman, Hastie, and Tibshirani 2008).”
BackgroundBelloni, A., Chen, M., & Chernozhukov, V. (2025) · Journal of Econometrics · doi:10.1016/j.jeconom.2025.106100 · Cited by 16
GGLasso - a Python package for General Graphical Lasso computation
Schaipp, F., Vlasovets, O., & Müller, C. L. (2021) · The Journal of Open Source Software · doi:10.21105/joss.03865 · Cited by 18
“The standard single Graphical Lasso problem (SGL) can be solved in sciki t-learn (Pedregosa et al. 2011).”
BackgroundMethodGreen Finance: Evidence from Large Portfolios and Networks during Financial Crises and Recessions
Argentiero, A., Bonaccolto, G., & Pedrini, G. (2022) · Corporate Social Responsibility and Environmental Management · doi:10.1002/csr.2687 · Cited by 13
Knowledge Discovery from Complex High Dimensional Data
Lee, S., & Holzinger, A. (2016) · Lecture notes in computer science · doi:10.1007/978-3-319-41706-6_7 · Cited by 16
“The sparse precision matrix estimation in (4) is a convex optimization problem proposed by Yuan and Lin [75].”
MethodA Poisson Log-Normal Model for Constructing Gene Covariation Network Using RNA-seq Data
Choi, Y., Coram, M., Peng, J., & Tang, H. (2017) · Journal of Computational Biology · doi:10.1089/cmb.2017.0053 · Cited by 16
“Gaussian graphical models (GGMs) offer an elegant framework to investigate the covarying patterns of many genes (Li and Gui, 2006; Meinshausen and Bu¨hlmann, 2006; Yuan and Lin, 2007; Peng et al., 2009).”
Dubart, M., Alonso, P., Barroso‐Bergadà, D., Becker, N., Béthune, K., Bohan, D. A., Boury, C., Cambon, M. C., Canard, E., Chancerel, É., et al. (2021) · Advances in ecological research/Advances in Ecological Research · doi:10.1016/bs.aecr.2021.10.007 · Cited by 14
Causal discoveries for high dimensional mixed data
Cai, Z., Xi, D., Zhu, X., & Li, R. (2022) · Statistics in Medicine · doi:10.1002/sim.9544 · Cited by 13
Sparse Inverse Covariance Matrices for Low Resource Speech Recognition
Zhang, W., & Fung, P. (2012) · IEEE Transactions on Audio Speech and Language Processing · doi:10.1109/tasl.2012.2221462 · Cited by 13
“They show that themethod can asymptotically recover the true graph.”
BackgroundMethodScheinberg, K., & Rish, I. (2009) · Cited by 13
Information theoretic structure learning with confidence
Moon, K. R., Noshad, M., Sekeh, S. Y., & Hero, A. O. (2017) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2017.7953327 · Cited by 13
“However, structure learning for GMs remains an open challenge since the most general case requires a combinatorial search over the space of all possible structures [8,9] and nonparametric approaches have poor convergence rates as the number of samples increases.”
BackgroundLearning Dynamic Conditional Gaussian Graphical Models
Huang, F., & Chen, S. (2017) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2017.2777462 · Cited by 13
“[7] and Banerjee et al.”
Nonparametric Finite Mixture of Gaussian Graphical Models
Lee, K. H., & Xue, L. (2017) · Technometrics · doi:10.1080/00401706.2017.1408497 · Cited by 16
“…and follow a parametric or semiparametric graphical model, for instance, Gaussian graphical model and its variants (Meinshausen and Bühlmann 2006; Yuan and Lin 2007; Friedman, Hastie, and Tibshirani 2008; Peng et al. 2009; Witten, Friedman, and Simon 2011; Cai, Liu, and Luo 2011; Liu et al.…”
MethodBayesian Inference in Nonparanormal Graphical Models.
Mulgrave, J. J. (2018) · NCSU Libraries Repository (North Carolina State University Libraries) · Cited by 13
Chen, X., Yang, D., Xu, Y., Yin, X., Wang, D., & Shen, H. (2021) · Journal of Econometrics · doi:10.1016/j.jeconom.2021.09.014 · Cited by 13
“Meanwhile, various methods of estimating the precision matrix have also been proposed (Meinshausen et al., 2006; Yuan and Lin, 2007; Friedman et al., 2008; Ravikumar et al., 2011; Cai et al., 2011, e.g.), and some other works extend the single precision matrix to multiple precision matrices…”
MethodEvolutionary Diagnosis of non-synonymous variants involved in differential drug response
Gerek, Z. N., Liu, L., Gerold, K., Biparva, P., Thomas, E. D., & Kumar, S. (2015) · BMC Medical Genomics · doi:10.1186/1755-8794-8-s1-s6 · Cited by 14
“Within a sparse-learning framework [37-40], we first performed a feature selection step to identify features with significant power to discriminate between DR-affecting (positive controls) and DR-neutral (negative controls) nsSNVs.”
MethodCai, T., & Yuan, M. (2015) · Journal of the American Statistical Association · doi:10.1080/01621459.2014.1001067 · Cited by 13
“See, for example, Ledoit and Wolf (2004), Huang et al. (2006), Yuan and Lin (2007), Bickel and Levina (2008a, b), El Karoui (2008), Fan, Fan, and Lv (2008), Friedman et al. (2008), Rothman et al. (2008), Lam and Fan (2009), Rothman, Levina, and Zhu (2009), Yuan (2010), Cai and Liu (2011), Cai, Liu,…”
BackgroundSelf-citationWang, B., Singh, R., & Qi, Y. (2017) · Machine Learning · doi:10.1007/s10994-017-5635-7 · Cited by 13
“e `1 -norm to create convex relaxation of the combinatorial formulation.For instance, the popular estimator graphical lasso (GLasso) has considered maximizing a `1 -penalizedlog-likelihoodobjective [63,2,23,55]. More recently, [5] proposed a constrained `1 -minimization formulation for estimating W , known as the CLIM…”
BackgroundMethodInformation-theoretic limits of graphical model selection in high dimensions
Santhanam, N., & Wainwright, M. J. (2008) · IEEE International Symposium on Information Theory (ISIT) · doi:10.1109/isit.2008.4595367 · Cited by 13
“Nonetheless, a variety of practical methods have been proposed, including constraint-based approaches [10], heuristic search, and l1-based relaxations [9], [12], [13].”
MethodA Pathwise Algorithm for Covariance Selection
Krishnamurthy, V., Ahipaşaoğlu, S. D., & d’Aspremont, A. (2011) · The MIT Press eBooks · doi:10.7551/mitpress/8996.003.0020 · Cited by 13
“This is a combinatorially hard (non-convex) problem and, as in [3, 4, 5, 6], we form the following convex relaxation maximize log detX − Tr(ΣX) − ρ‖X‖1 (1) which is a convex problem in the matrix variable X ∈ Sn, where ‖X‖1 is the sum of absolute values of the coefficients of X here.”
BackgroundLearning Scale-Free Networks by Dynamic Node-Specific Degree PriorPreprint
Tang, Q., Sun, S., & Xu, J. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1503.02129 · Cited by 13
“Estimating the structure of an undirected graphical model from a dataset has been extensively studied (Meinshausen & Bühlmann, 2006; Yuan & Lin, 2007; Friedman et al., 2008; Banerjee et al., 2008; Wainwright et al., 2006).”
MethodWang, Y., Roy, U., & Uhler, C. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1906.05159 · Cited by 13
“The most prominent of these algorithms is graphical lasso (Banerjee et al., 2008; Friedman et al., 2008; Ravikumar et al., 2011; Yuan and Lin, 2007), which applies an `1 penalty to the log-likelihood function to estimate the precision matrix.”
MethodOverlapping decomposition for causal graphical modeling
Han, L., Song, G., Cong, G., & Xie, K. (2012) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/2339530.2339551 · Cited by 12
“This minimization problem can be solved efficiently by the graphical lasso algorithms proposed in [4, 19].”
BackgroundMethodInfluentialA comparative study of covariance selection models for the inference of gene regulatory networks
Stifanelli, P. F., Creanza, T. M., Anglani, R., Liuzzi, V. C., Mukherjee, S., Schena, F. P., & Ancona, N. (2013) · Journal of Biomedical Informatics · doi:10.1016/j.jbi.2013.07.002 · Cited by 16
“Finally, the authors in [34] suggested an attractive and simple approach based on lasso-type regression to select the non-vanishing partial correlations, paving the way to a number of analysis and novel algorithms based on lasso ‘ 1 regulariza-tions [50,17,49,18].”
MethodForward Basis Selection for Pursuing Sparse Representations over a Dictionary
Yuan, X., & Yan, S. (2013) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2013.85 · Cited by 12
“Given a sample covariance matrix n, Yuan and Lin [38] formulated sparse inverse covariance estimation as the following ‘1-norm penalized log-determinant program: min 2SSpþ log det þ h n; i þ j j1: ð31Þ There exist a variety of optimization algorithms addressing this convex formulation [3], [11], [30], [28], [23], [37]…”
MethodYu, D., Son, W., Lim, J., & Xiao, G. (2015) · Biostatistics · doi:10.1093/biostatistics/kxv013 · Cited by 12
“Yuan and Lin (2007) propose the 1-regularized maximum likelihood estimator (MLE) of the concentration matrix and an algorithm to solve it using the determinant maximization problem (MAXDET), which has computational complexity of order O(p6) and becomes slower as p increases.”
MethodLiu, Z., Sun, F., Braun, J. G., McGovern, D. P. B., & Piantadosi, S. (2014) · Bioinformatics · doi:10.1093/bioinformatics/btu778 · Cited by 15
“Among all the methods, various graphical models with L1 regularization are perhaps the most common approach for graphical structure estimation (Banerjee et al., 2008; Liu and Ihler, 2011; Meinshausen and Bühlmann, 2006; Peng et al., 2009; Yuan and Lin, 2007).”
MethodLi, J., & Zhong, P. (2017) · The Annals of Statistics · doi:10.1214/16-aos1459 · Cited by 15
On High-Dimensional Constrained Maximum Likelihood Inference
Zhu, Y., Shen, X., & Pan, W. (2018) · Journal of the American Statistical Association · doi:10.1080/01621459.2018.1540986 · Cited by 12
“For the Gaussian graphical model, methods include the regularized likelihood approach (Rothman et al. 2008; Friedman, Hastie, and Tibshirani 2008; Yuan and Lin 2007; Fan, Feng, and Wu 2009; Shen, Pan, and Zhu 2012) and the nodewise CONTACT Yunzhang Zhu zhu.219@osu.edu Department of Statistics, Ohio…”
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.