Model selection and estimation in the Gaussian graphical model Export

@article{yuan07, 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. 10.1093/biomet/asm018}, author = {Yuan, Ming and Lin, Yi}, citeulike-article-id = {1160802}, citeulike-linkout-0 = {http://dx.doi.org/10.1093/biomet/asm018}, citeulike-linkout-1 = {http://biomet.oxfordjournals.org/cgi/content/abstract/94/1/19}, citeulike-linkout-2 = {http://www.ingentaconnect.com/content/oup/biomet/2007/00000094/00000001/art00019}, day = {1}, doi = {10.1093/biomet/asm018}, issn = {0006-3444}, journal = {Biometrika}, keywords = {covariance, glasso, inverse, model, selection}, month = {March}, number = {1}, pages = {19--35}, posted-at = {2009-10-30 19:53:22}, priority = {0}, publisher = {Oxford University Press}, title = {Model selection and estimation in the Gaussian graphical model}, url = {http://dx.doi.org/10.1093/biomet/asm018}, volume = {94}, year = {2007} }

TY - JOUR ID - yuan07 L3 - citeulike-article-id:1160802 N2 - 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. 10.1093/biomet/asm018 IS - 1 JF - Biometrika SN - 0006-3444 EP - 35 TI - Model selection and estimation in the Gaussian graphical model PB - Oxford University Press VL - 94 SP - 19 KW - covariance KW - glasso KW - inverse KW - model KW - selection AU - Yuan, Ming AU - Lin, Yi PY - 2007/03/01/ UR - http://dx.doi.org/10.1093/biomet/asm018 ER -