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 701–800 of 1,000
Learning structured densities via infinite dimensional exponential families
Sun, S., Kolar, M., & Xu, J. (2015) · neural information processing systems · Cited by 9
Graphical models via joint quantile regression with component selection
Chun, H., Lee, M. H., Fleet, J. C., & Oh, J. H. (2016) · Journal of Multivariate Analysis · doi:10.1016/j.jmva.2016.07.012 · Cited by 9
Graphical tool of sparse factor analysis
Yamamoto, M., Hirose, K., & Nagata, H. (2017) · Behaviormetrika · doi:10.1007/s41237-016-0007-3 · Cited by 9
“Over the last two decades, various techniques to enhance sparsity have been developed in several statistical problems, such as regression, graphical model, and principal component analysis (Tibshirani 1996; Yuan and Lin 2007; Zou 2006).”
MethodShrinking characteristics of precision matrix estimators
Molstad, A. J., & Rothman, A. J. (2018) · Biometrika · doi:10.1093/biomet/asy023 · Cited by 9
“…tuning parameter chosen to minimize misclassification rate on the validation set; the L 1 - penalized Gaussian likelihood precision matrix estimator (Yuan and Lin, 2007; Rothman et al., 2008; Friedman et al., 2008) with the tuning parameter chosen to minimize the misclassification rate of the…”
BackgroundMethodBayesian Lasso with neighborhood regression method for Gaussian graphical model
Fan-qun, L., & Zhang, X. (2017) · Acta Mathematicae Applicatae Sinica English Series · doi:10.1007/s10255-017-0676-z · Cited by 9
“Recently, the graphical Lasso is not only used for graphical model selection (Meinshausen and Buhlmann [13] ), but also widely used for simultaneous graphical model determination and parameter estimation (Yuan and Lin [24] ; Banerjee et al. [4] ; Fridman et al. [9] ).”
MethodHe, S., & Deng, M. (2019) · PLoS ONE · doi:10.1371/journal.pone.0207731 · Cited by 12
Structure Learning of Gaussian Markov Random Fields with False Discovery Rate Control
Lee, S., Sobczyk, P., & Bogdan, M. (2019) · Symmetry · doi:10.3390/sym11101311 · Cited by 9
Sparsity Inducing Prior Distributions for Correlation Matrices of Longitudinal Data
Gaskins, J., Daniels, M. J., & Marcus, B. H. (2013) · Journal of Computational and Graphical Statistics · doi:10.1080/10618600.2013.852553 · Cited by 11
Map approach to learning sparse Gaussian Markov networks
Asadi, N. B., Rish, I., Scheinberg, K., Kanevsky, D., & Ramabhadran, B. (2009) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2009.4959935 · Cited by 9
Non-convex Statistical Optimization for Sparse Tensor Graphical Model.
Sun, W., Wang, Z., Liu, H., & Cheng, G. (2015) · PubMed · Cited by 9
Fast Sparse Gaussian Markov Random Fields Learning Based on Cholesky Factorization
Stojković, I., Jelisavčić, V., Milutinović, V. M., & Zoran Obradovic, A. (2017) · International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2017/384 · Cited by 9
“Previous work [Yuan and Lin, 2007; Banerjee et al., 2008; Friedman et al., 2008] relies on applying the L1 penalty on the elements of a precision matrix.”
MethodFast learning of scale‐free networks based on Cholesky factorization
Jelisavčić, V., Stojković, I., Milutinović, V. M., & Zoran Obradovic, A. (2018) · International Journal of Intelligent Systems · doi:10.1002/int.21984 · Cited by 9
“Initial approaches 15 were computationally expensive, and suitable for only small problems.”
BackgroundBanerjee, S., Akbani, R., & Baladandayuthapani, V. (2018) · Computational Statistics & Data Analysis · doi:10.1016/j.csda.2018.08.009 · Cited by 9
Gaussian Graphical Model Exploration and Selection in High Dimension Low Sample Size Setting
Lartigue, T., Bottani, S., Baron, S., Olivier Colliot, A., Durrleman, S., & Allassonnière, S. (2020) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2020.2980542 · Cited by 9
“In the latter, the structure learning steps are a variation of the parallel nodewise linear regressions of [1] and the model selection steps guided by out of sample versions of the likelihood optimised in [3] and [4].”
BackgroundMethodInfluentialESPN: Extremely Sparse Pruned Networks
Cho, M., Joshi, A., & Hegde, C. (2021) · IEEE Data Science Workshop (DSW) · doi:10.1109/dslw51110.2021.9523404 · Cited by 11
Zhang, Y., Tao, J., Yin, Z., & Wang, G. (2022) · Mathematics · doi:10.3390/math10224282 · Cited by 9
Common Substructure Learning of Multiple Graphical Gaussian Models
Hara, S., & Washio, T. (2011) · Lecture notes in computer science · doi:10.1007/978-3-642-23783-6_1 · Cited by 9
“Though classical approaches have encountered several difficulties, there is a recent development on the use of 1-regularization [8,9,10], that enables the design of an efficient Graphical Lasso (GLasso) algorithm [11].”
MethodInfluentialInferring Gene Network Rewiring by Combining Gene Expression and Gene Mutation Data
Tu, J.-J., Ou-Yang, L., Hu, X. T., & Zhang, X.-F. (2018) · IEEE Transactions on Computational Biology and Bioinformatics · doi:10.1109/tcbb.2018.2834529 · Cited by 9
“Therefore, the state-specific gene networks can be modeled by the corresponding precision matrices [12], [13], [14], and the differential network between two states can be modeled as the difference between the two state-specific precision matrices [4], [15], [16], [17], [18].”
BackgroundInference for elliptical copula multivariate response regression models
Zhao, Y., & Genest, C. (2019) · Electronic Journal of Statistics · doi:10.1214/19-ejs1534 · Cited by 9
“Then, to deduce an estimator of Ωεε from the estimator of Σεε, we adopt the approach of [33, 34, 45].”
Koanantakool, P., Ali, A., Azad, A., Buluç, A., Morozov, D., Oliker, L., Yelick, K., & Oh, S. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1710.10769 · Cited by 9
A two-way factor model for high-dimensional matrix dataPreprint
Gao, Z., Chaofeng, Y., Jing, B., Wei, H., & Guo, J. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2103.07920 · Cited by 9
“…several methods have been suggested for when this situation occurs (Sheena and Gupta, 2003; Vandenberghe and Boyd, 2004; Wainwright et al, 2006; Yuan and Lin, 2007; Yuan, 2009), they are only designed to handle problems, and most of them essentially follow indirect ways to handle the relevant…”
MethodThe multivariate normal distribution
Lauritzen, S. L. (1996) · Oxford University Press eBooks · doi:10.1093/9780191775284.005.0003 · Cited by 9
Covariance Model with General Linear Structure and Divergent Parameters
Fan, X., Lan, W., Zou, T., & Tsai, C. (2022) · Journal of Business and Economic Statistics · doi:10.1080/07350015.2022.2142593 · Cited by 8
“…role in modern multivariate analysis and its applications across various fields such as risk management, portfolio allocation, biostatistics, social networks and health sciences (see, e.g., Kan and Zhou 2007; Friedman, Hastie, and Tibshirani 2008; Yuan and Lin 2007; Fan, Han, and Liu 2014).”
BackgroundGraphical lassos for meta‐elliptical distributions
Bilodeau, M. (2014) · Canadian Journal of Statistics · doi:10.1002/cjs.11212 · Cited by 8
“Yuan & Lin (2007) showed that this simple approach can be viewed as an approximation to the exact maximization of the 1 penalized likelihood.”
BackgroundMethodHierarchical array priors for ANOVA decompositions of cross-classified data
Volfovsky, A., & Hoff, P. D. (2014) · The Annals of Applied Statistics · doi:10.1214/13-aoas685 · Cited by 8
“Recent variants of these approaches allow for different penalties on ANOVA terms of different orders, including the ASP method of Beran (2005) and grouped versions of the lasso [Yuan and Lin (2007), Friedman, Hastie and Tibshirani (2010)].”
Algorithms for Sparse and Low-Rank Optimization: Convergence, Complexity and Applications
Ma, S. (2011) · doi:10.7916/d8pc38bz · Cited by 11
“, [3, 39, 102, 107]).”
Incorporating Graphical Structure of Predictors in Sparse Quantile Regression
Wang, Z., Liu, X., Tang, W., & Lin, Y. (2020) · Journal of Business and Economic Statistics · doi:10.1080/07350015.2020.1730859 · Cited by 8
“Such a predictor graph is readily available from prior information (Subramanian et al. 2005) or statistical estimation of the covariance matrix or precision matrix (see Yuan and Lin 2007; Friedman, Hastie, and Tibshirani 2008; Cai, Liu, and Luo 2011).”
MethodSelf-citationLearnable Graph-Regularization for Matrix Decomposition
Zhai, P., & Zhang, S. (2023) · ACM Transactions on Knowledge Discovery from Data · doi:10.1145/3544781 · Cited by 8
“This problem has been extensively studied and many algorithms have been proposed [5, 9, 18, 24, 27, 49].”
BackgroundInter-Subject Analysis: A Partial Gaussian Graphical Model Approach
Ma, C., Lu, J., & Liu, H. (2020) · Journal of the American Statistical Association · doi:10.1080/01621459.2020.1841645 · Cited by 10
“Estimators for the precision matrix including the maximum likelihood estimator and the column-wise estimator have been considered in Yuan and Lin (2007), Banerjee, Ghaoui, and d’Aspremont (2008), Rothman et al. (2008), Friedman, Hastie, and Tibshirani (2008), Ravikumar et al. (2011), Meinshausen…”
MethodWu, Z., Wang, C., & Liu, W. (2022) · Annals of the Institute of Statistical Mathematics · doi:10.1007/s10463-022-00856-0 · Cited by 8
“…al. (2012) proposed four precision matrix estimators by plugging the rank-based estimate into the Dantzig-type method (Yuan, 2010), the CLIME (Cai et al., 2011), the graphical Lasso (Yuan and Lin, 2007; Friedman et al., 2008) and the neighbourhood pursuit estimator(Meinshausen and B¨uhlmann, 2006).”
MethodAgoalikum, E., Klugah‐Brown, B., Wu, H., Jing, J., & Biswal, B. B. (2023) · Brain Connectivity · doi:10.1089/brain.2022.0069 · Cited by 8
“Kindly refer to Banerjee et al, 2008; Meinshausen and Bu¨hlmann, 2006; Yuan and Lin, 2007 for more information.”
Learning graph Laplacian with MCP
Zhang, Y., Toh, K.-C., & Sun, D. (2023) · Optimization methods & software · doi:10.1080/10556788.2023.2269594 · Cited by 8
“:3)) and its sample covariance matrix S ∈ S n + , a notable way of learning a precision matrix from the data matrix S is via the following (cid:4) 1 -norm penalized maximum likelihood approach [4,20,46,63] min where 1,off = i (cid:7)= j | (cid:5) ij | and λ is a non-negative penalty parameter.”
InfluentialNouri, N., Gaglia, G., Mattoo, H., de Rinaldis, E., & Savova, V. (2024) · Cell Reports Methods · doi:10.1016/j.crmeth.2024.100794 · Cited by 8
On model misspecification and KL separation for Gaussian graphical models
Jog, V., & Loh, P. (2015) · IEEE International Symposium on Information Theory (ISIT) · doi:10.1109/isit.2015.7282640 · Cited by 8
“Recent years have also seen substantial theoretical advances regarding graphical models in high-dimensional statistics (see, e.g., [6], [7], [8], [9], [10]).”
BackgroundOptimal estimation of sparse correlation matrices of semiparametric Gaussian copulas
Xue, L., & Zou, H. (2014) · Statistics and Its Interface · doi:10.4310/sii.2014.v7.n2.a5 · Cited by 8
Sparse Gaussian graphical mixture model
Lotsi, A., & Wit, E. C. (2016) · Afrika Statistika · doi:10.16929/as/2016.1041.91 · Cited by 8
Multiclass Vector Auto-Regressive Models for Multistore Sales Data
Wilms, I., Barbaglia, L., & Croux, C. (2017) · Journal of the Royal Statistical Society Series C (Applied Statistics) · doi:10.1111/rssc.12231 · Cited by 8
“The optimal values of the regularization parameters γ1 and γ2 are selected using the BIC (e.g. Yuan and Lin, 2007).”
DIFFERENTIALLY PRIVATE SPARSE INVERSE COVARIANCE ESTIMATION
Wang, D., Huai, M., & Xu, J. (2018) · IEEE Global Conference on Signal and Information Processing · doi:10.1109/globalsip.2018.8646444 · Cited by 8
“Estimating the inverse covariance matrix (also called precision matrix) in high dimensional space is a fundamental problem in statistics and finds applications in many fields such as machine leaning, signal processing, computational biology, etc [21] [10].”
BackgroundDas, A., Sexton, D., Lainscsek, C. S. M., Cash, S. S., & B, T. J. S. A. (2019) · Neural Computation · doi:10.1162/neco_a_01205 · Cited by 8
“…et al., 2008; Hsieh, Sustik, Dhillon, & Ravikumar, 2011; Hsieh, Sustik, Dhillon, Ravikumar, & Poldrack, 2013; Oztoprak, Nocedal, Rennie, & Olsen, 2012; Rothman, Bickel, Levina, & Zhu, 2008; Scheinberg, Ma, & Goldfarb, 2010; Yuan & Lin, 2007) have been proposed for solving this optimization problem.”
Wang, J., Westveld, A. H., Welsh, A. H., Parker, M. A., & Loong, B. (2021) · PLoS ONE · doi:10.1371/journal.pone.0248340 · Cited by 8
Unveiling Anomalous Edges and Nominal Connectivity of Attributed Networks
Polyzos, K. D., Mavromatis, C., Ioannidis, V. N., & Giannakis, G. B. (2020) · Asilomar Conference on Signals, Systems and Computers (ACSSC) · doi:10.1109/ieeeconf51394.2020.9443464 · Cited by 8
“Popular approaches to estimating graph structure are closely related to the sparse inverse covariance matrix estimation [32], [33], [34].”
BackgroundSparse Laplacian Shrinkage with the Graphical Lasso Estimator for Regression Problems
Xia, S., Yang, Y., & Yang, H. (2021) · Test · doi:10.1007/s11749-021-00779-7 · Cited by 8
Ghosh, I. (2018) · Handbook of statistics · doi:10.1016/bs.host.2018.07.001 · Cited by 8
Discriminative Mixtures of Sparse Latent Fields for Risk Management
Agakov, F. V., Orchard, P., & Storkey, A. (2012) · International Conference on Artificial Intelligence and Statistics · Cited by 9
Tests for mean vectors in high dimension
Maboudou‐Tchao, E. M., & Silva, I. R. (2013) · Statistical Analysis and Data Mining The ASA Data Science Journal · doi:10.1002/sam.11209 · Cited by 8
“In the absence of model assumptions when P > N , an active line of statistical research is based on imposing various restrictions on the model—for instance, sparsity [14,16,17,23,24].”
BackgroundA joint convex penalty for inverse covariance matrix estimation
Maurya, A. (2014) · Computational Statistics & Data Analysis · doi:10.1016/j.csda.2014.01.015 · Cited by 8
TEMPORAL GRAPHICAL MODELS FOR CROSS-SPECIES GENE REGULATORY NETWORK DISCOVERY
LIU, Y., Niculescu-Mizil, A., Lozano, A. C., & Lu, Y. C. (2011) · Journal of Bioinformatics and Computational Biology · doi:10.1142/s0219720011005525 · Cited by 9
“each task) are generated from Gaussian distributions using examples of the AR models used in [43, 18] (notice that this corresponds to the gene-regulatory networks of individual species in the application of cross-species regulatory network discovery).”
BackgroundMethodA Dynamic Bayesian Model for Characterizing Cross-Neuronal Interactions During Decision-Making
Zhou, B., Moorman, D. E., Behseta, S., Ombao, H., & Shahbaba, B. (2015) · Journal of the American Statistical Association · doi:10.1080/01621459.2015.1116988 · Cited by 9
Mixed Scale Joint Graphical Lasso
Pircalabelu, E., Claeskens, G., & Waldorp, L. (2016) · Biostatistics · doi:10.1093/biostatistics/kxw025 · Cited by 8
Using network analysis to study behavioural phenotypes: an example using domestic dogs
Goold, C. M., Vas, J., Olsen, C., & Newberry, R. C. (2016) · Royal Society Open Science · doi:10.1098/rsos.160268 · Cited by 9
“To construct networks that represented conditional independence relationships, we used Gaussian graphical models (GGM; see [55,56] for an overview).”
MethodFitting Exploratory Factor Analysis Models with High Dimensional Psychological Data
Finch, W. H., & Finch, M. E. H. (2021) · Journal of Data Science · doi:10.6339/jds.201607_14(3).0008 · Cited by 8
Recent developments in high dimensional covariance estimation and its related issues, a review
Hong, Y., & Kim, C. (2018) · Journal of the Korean Statistical Society · doi:10.1016/j.jkss.2018.04.005 · Cited by 8
He, Y., Ji, J., Xie, L., Zhang, X., & Xue, F. (2018) · BMC Bioinformatics · doi:10.1186/s12859-018-2461-2 · Cited by 8
De-Biased Graphical Lasso for High-Frequency Data
Koike, Y. (2020) · Entropy · doi:10.3390/e22040456 · Cited by 8
“Following [17,55], we select it by minimizing the following formally defined Bayesian information criterion (BIC):”
MethodRobust and sparse Gaussian graphical modelling under cell‐wise contamination
Katayama, S., Fujisawa, H., & Drton, M. (2018) · Stat · doi:10.1002/sta4.181 · Cited by 8
“Yuan & Lin (2007), Banerjee et al. (2008) and Friedman et al. (2008) considered the graphical lasso (Glasso) that involves the ℓ 1 penalized log-likelihood function.”
MethodHeterogeneity adjustment with applications to graphical model inference
Fan, J., Liu, H., Wang, W., & Zhu, Z. (2018) · Electronic Journal of Statistics · doi:10.1214/18-ejs1466 · Cited by 11
“[49] and [20] developed the Graphical Lasso method using the L1 penalty and [27] and [42] used a non-convex penalty.”
MethodA Latent Gaussian Copula Model for Mixed Data Analysis in Brain Imaging Genetics
Zhang, A., Fang, J., Hu, W., Calhoun, V. D., & Wang, Y. (2019) · IEEE Transactions on Computational Biology and Bioinformatics · doi:10.1109/tcbb.2019.2950904 · Cited by 10
“Thus, we compared the proposed latent correlation with Pearson correlation using gLASSO ([18]) on the MCIC study with fMRI image and SNP data.”
MethodInfluentialMinimax estimation of large precision matrices with bandable Cholesky factor
Liu, Y., & Ren, Z. (2020) · The Annals of Statistics · doi:10.1214/19-aos1893 · Cited by 8
“6 LIU AND REN likelihood estimation ([27, 57, 17, 2, 45, 32, 43]) and penalized regression methods ([38, 56, 16, 49, 44]) are designed for sparse precision matrix estimation.”
MethodEstimating Multiple Precision Matrices With Cluster Fusion Regularization
Price, B. S., Molstad, A. J., & Sherwood, B. (2021) · Journal of Computational and Graphical Statistics · doi:10.1080/10618600.2021.1874963 · Cited by 8
“Sparsity inducing penalties are especially popular in penalized normal maximum likelihood (Yuan and Lin 2007; d’Aspermont, Banerjee, and Ghaoui 2008; Friedman, Hastie, and Tibshirani 2007; Rothman et al. 2008; Witten, Friedman, and Simon 2011) because a zero in the (j, k)th entry of ∗c implies the…”
BackgroundA Flexible, Scalable and Efficient Algorithmic Framework for Primal Graphical LassoPreprint
Mazumder, R., & Agarwal, D. (2011) · arXiv (Cornell University) · doi:10.48550/arxiv.1110.5508 · Cited by 8
“The ℓ1 regularization (Friedman et al., 2007b; Banerjee et al., 2008; Yuan & Lin, 2007; Meinshausen & Bühlmann, 2006) is often used in this context since it performs smoothing, induces sparsity, adds interpretation and forms an effective model selection procedure.”
MethodPANDA: AdaPtive Noisy Data Augmentation for Regularization of Undirected Graphical ModelsPreprint
Li, Y., Liu, X., & Liu, F. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1810.04851 · Cited by 8
Estimation of (near) low-rank matrices with noise and high-dimensional scalingPreprint
Negahban, S. N., & Wainwright, M. J. (2009) · arXiv (Cornell University) · Cited by 8
B-CONCORD -- A scalable Bayesian high-dimensional precision matrix estimation procedurePreprint
Jalali, P., Khare, K., & Michailidis, G. C. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2005.09017 · Cited by 9
“…estimation of a sparse precision matrix, either based on the neigborhood selection idea or leveraging the Gaussian likelihood; e.g., see Meinshausen and Buhlmann [2006], Yuan and Lin [2007], Fried-man et al. [2008], Peng et al. [2009], Cai et al. [2011], Khare et al. [2015] and references therein.”
MethodJoint estimation for multisource Gaussian graphical models based on transfer learning
Zhang, Y., & Yang, Y. (2024) · Pattern Recognition · doi:10.1016/j.patcog.2024.110964 · Cited by 7
Discussion: Latent variable graphical model selection via convex optimization
Lauritzen, S. L., & Meinshausen, N. (2012) · The Annals of Statistics · doi:10.1214/12-aos980 · Cited by 7
“It has been framed as a computationally convenient convex ℓ1-constrained estimation problem in Yuan and Lin (2007) and can be applied readily to higher-dimensional problems.”
MethodA coordinate descent algorithm for sparse positive definite matrix estimation
Yuan, T., & Wang, J. (2013) · Statistical Analysis and Data Mining The ASA Data Science Journal · doi:10.1002/sam.11185 · Cited by 7
“(1) with the lasso penalty J ( ) = ‖ ‖1 = ∑p j,k=1 | jk| or J ( ) = ‖ −‖1 = ∑ j =k | jk|, including the max-det optimization algorithm [6], the graphical lasso [4], and the Cholesky decomposition [7].”
MethodUniform inference in high-dimensional Gaussian graphical models
Klaaßen, S., Kueck, J., Spindler, M., & Chernozhukov, V. (2022) · Biometrika · doi:10.1093/biomet/asac030 · Cited by 10
Chung, H. C., Gaynanova, I., & Ni, Y. (2022) · The Annals of Applied Statistics · doi:10.1214/21-aoas1598 · Cited by 10
“Yuan and Lin (2007); Banerjee et al. (2008); Dahl et al. (2008); Friedman et al. (2008) directly estimate the sparse concentration matrix by optimizing the L1penalized log-likelihood function, the so-called graphical lasso.”
Correlation Imputation for Single-Cell RNA-seq
Gan, L., Vinci, G., & Allen, G. I. (2022) · Journal of Computational Biology · doi:10.1089/cmb.2021.0403 · Cited by 9
“Specifically, we plug the correlation matrix estimates into the graphical lasso (Yuan and Lin, 2007) to obtain a gene-by-gene network estimate through sparse precision matrix estimation.”
InfluentialSeal, S., Li, Q., Basner, E. B., Saba, L. M., & Kechris, K. J. (2023) · PLoS Computational Biology · doi:10.1371/journal.pcbi.1010758 · Cited by 7
“Yuan and Lin [16], Banerjee et al. [17], Friedman et al. [18], considered a penalized maximum likelihood model with (cid:96) 1 regularization, known as graphical lasso (GL).”
BackgroundDiscussion: Latent variable graphical model selection via convex optimization
Candès, E. J., & Soltanolkotabi, M. (2012) · The Annals of Statistics · doi:10.1214/12-aos1001 · Cited by 7
“The proposed scheme is an extension of the graphical lasso of Yuan and Lin [15] (see also [1, 6]), which is a popular approach for learning the structure in an undirected Gaussian graphical model.”
Sparse networks of directly coupled, polymorphic, and functional side chains in allosteric proteins
Ghoraie, L. S., Burkowski, F. J., & Zhu, M. (2014) · Proteins Structure Function and Bioinformatics · doi:10.1002/prot.24752 · Cited by 8
Adaptive Sparsity in Gaussian Graphical Models
Wong, E., Awate, S. P., & Fletcher, P. T. (2013) · International Conference on Machine Learning · Cited by 7
Lozano, A. C., Jiang, H., & Deng, X. (2013) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/2487575.2487667 · Cited by 7
“Minimum distance estimators [33] are popularized in nonparametric methods and have exhibited excellent robustness properties [3, 12].”
BackgroundIn praise of sparsity and convexity
Tibshirani, R. J. (2013) · Cited by 7
Manifold Regularized Experimental Design for Active Learning
Zhang, L., Shum, H. P. H., & Shao, L. (2016) · IEEE Transactions on Image Processing · doi:10.1109/tip.2016.2635440 · Cited by 7
“As shown in [53], the ‖β‖1 results in a sparse coefficient β.”
Yuan, R., Ou-Yang, L., Hu, X. T., & Zhang, X.-F. (2019) · IEEE Transactions on Computational Biology and Bioinformatics · doi:10.1109/tcbb.2019.2901473 · Cited by 7
“Two genes are conditional dependent if and only if the corresponding element in the inverse covariance matrix is zero [16].”
BackgroundLi, M., Wu, Y., He, Y., Huang, S., & Nair, A. (2019) · Decision Sciences · doi:10.1111/deci.12404 · Cited by 7
Estimation of Dynamic Networks for High-Dimensional Nonstationary Time Series
Xu, M., Chen, X., & Wu, W. B. (2019) · Entropy · doi:10.3390/e22010055 · Cited by 7
“On one hand, there is a large volume of literature on estimating the (static) precision matrix for graphical models in the high-dimensional setting, where the sample size and the dimension are both large [44, 27, 3, 53, 61, 62, 52, 11, 10, 9, 25, 6, 40, 41].”
BackgroundLiu, T., Trinh-Hoang, M., Yang, Y., & Pesavento, M. (2019) · IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing · doi:10.1109/camsap45676.2019.9022643 · Cited by 7
“To overcome this challenge and also to encourage a sparse graph, the graphical LASSO estimator, which regularizes the ML estimator with the sparsity-promoting (cid:96) 1 -norm, has been proposed in [1].”
BackgroundMethodInfluentialMicrobial Interaction Network Estimation via Bias-Corrected Graphical Lasso
Jiang, D., Sharpton, T. J., & Jiang, Y. (2020) · Statistics in Biosciences · doi:10.1007/s12561-020-09279-y · Cited by 7
Sparsity-aware robust community detection (SPARCODE)
Taştan, A., Muma, M., & Zoubir, A. M. (2021) · Signal Processing · doi:10.1016/j.sigpro.2021.108147 · Cited by 8
Bayesian inference for high-dimensional decomposable graphs
Lee, K., & Cao, X. (2021) · Electronic Journal of Statistics · doi:10.1214/21-ejs1822 · Cited by 7
“For all the penalized likelihood methods (Friedman, Hastie and Tibshirani, 2007; Cai, Liu and Luo, 2011; Yuan and Lin, 2007), a user-specified penalty parameter controls the level of sparsity of the resulting estimator.”
MethodOmolofe, O. T., Adegoke, N. A., Adeoti, O. A., Fasoranbaku, O. A., & Abbasi, S. A. (2021) · Quality Technology & Quantitative Management · doi:10.1080/16843703.2021.1948952 · Cited by 7
Monitoring multivariate coefficient of variation for high‐dimensional processes
Adegoke, N. A., Dawod, A. B. A., Adeoti, O. A., Sanusi, R. A., & Abbasi, S. A. (2022) · Quality and Reliability Engineering International · doi:10.1002/qre.3094 · Cited by 7
Liu, D., Zhao, C., He, Y., Liu, L., Guo, Y., & Zhang, X. (2022) · Biometrics · doi:10.1111/biom.13753 · Cited by 7
Katayama, S., & Imori, S. (2014) · Journal of Multivariate Analysis · doi:10.1016/j.jmva.2014.08.002 · Cited by 7
Hypothesis testing for band size detection of high-dimensional banded precision matrices
An, B., Guo, J., & Liu, Y. (2014) · Biometrika · doi:10.1093/biomet/asu006 · Cited by 7
“One popular method for precision matrix estimation is to maximize the Gaussian likelihood with a penalty on the precision matrix (Meinshausen & Buhlmann, 2006; Yuan & Lin, 2007; Friedman et al., 2008).”
MethodAnalyzing and learning sparse and scale-free networks using Gaussian graphical models
Aslan, M. S., Chen, X.-W., & Cheng, H. (2016) · International Journal of Data Science and Analytics · doi:10.1007/s41060-016-0009-y · Cited by 7
“[12] proposed a penalized-likelihood method which has the model selection and parameter estimation simultaneously in the precision matrix estimation.”
MethodEstimation of Sparse Directed Acyclic Graphs for Multivariate Counts Data
Han, S. W., & Zhong, J. (2016) · Biometrics · doi:10.1111/biom.12467 · Cited by 7
“For a predetermined integer K , one can employ a K -fold cross-validation (CV) method to select the best tuning parameter in the proposed plnDAG model by criteria such as minimizing the prediction error or the Bayesian information criterion (BIC) (Yuan and Lin, 2007; Rothman et al., 2008).”
MethodConic Optimization Theory: Convexification Techniques and Numerical Algorithms
Zhang, R. Y., Josz, C., & Sojoudi, S. (2018) · Proceedings of the ... American Control Conference/Proceedings of the American Control Conference · doi:10.23919/acc.2018.8430887 · Cited by 7
Inference of high-dimensional linear models with time-varying coefficients
Chen, X., & He, Y. (2017) · Statistica Sinica · doi:10.5705/ss.202015.0202 · Cited by 12
On generating random Gaussian graphical models
Córdoba, I., Varando, G., Bielza, C., & Larrañaga, P. (2020) · International Journal of Approximate Reasoning · doi:10.1016/j.ijar.2020.07.007 · Cited by 7
“Looking 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.”
BackgroundThe conditional censored graphical lasso estimator
Augugliaro, L., Sottile, G., & Vinciotti, V. (2020) · Statistics and Computing · doi:10.1007/s11222-020-09945-7 · Cited by 7
Joint estimation of multiple mixed graphical models for pan‐cancer network analysis
Jia, B., & Liang, F. (2020) · Stat · doi:10.1002/sta4.271 · Cited by 7
Post-selection inference of generalized linear models based on the lasso and the elastic net
Shi, X., Liang, B., & Zhang, Q. (2020) · Communication in Statistics- Theory and Methods · doi:10.1080/03610926.2020.1821892 · Cited by 8
“To improve the prediction accuracy and explainability of the model, a large number of variable selection methods have emerged, such as lasso (Tibshirani 1996), SCAD (Fan and Li 2001), elastic net (Zou and Hastie 2005), adaptive lasso (Zou 2006) and group lasso (Yuan and Lin 2007), etc.”
MethodEnsemble sparse estimation of covariance structure for exploring genetic disease data
Kang, X., & Wang, M. (2021) · Computational Statistics & Data Analysis · doi:10.1016/j.csda.2021.107220 · Cited by 7
Joint Association Graph Screening and Decomposition for Large-Scale Linear Dynamical Systems
She, Y., He, Y., Li, S., & Wu, D. O. (2014) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2014.2373315 · Cited by 7
“Following [25, 46, 47], we chose the tuning parameters”
BackgroundMethodBayesian Adjustment for Multiplicity
Scott, J. G. (2009) · DukeSpace (Duke University) · Cited by 9
Maurya, A. (2014) · arXiv (Cornell University) · doi:10.48550/arxiv.1412.7907 · Cited by 7
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.