This paper considers estimating a covariance matrix of p variables from n observations by either banding the sample covariance matrix or estimating a banded version of the inverse of the covariance. We show that these estimates are consistent in the operat
Regularizedestimationoflargecovariancematrices
PeterJ.BickelandElizavetaLevina
UniversityofCalifornia,BerkeleyandUniversityofMichigan
September14,2006
Abstract
Thispaperconsidersestimatingacovariancematrixofpvariablesfromnobservationsbyeitherbandingthesamplecovariancematrixorestimatingabandedversionoftheinverseofthecovariance.Weshowthattheseestimatesareconsistentintheoperatornormaslongas(logp)2/n→0,andobtainexplicitrates.Theresultsareuniformoversomefairlynaturalwell-conditionedfamiliesofcovariancematrices.WealsointroduceananalogueoftheGaussianwhitenoisemodelandshowthatifthepopulationcovarianceisembeddableinthatmodelandwell-conditionedthenthebandedapproximationsproduceconsistentestimatesoftheeigen-valuesandassociatedeigenvectorsofthecovariancematrix.Theresultscanbeextendedtosmoothversionsofbandingandtonon-Gaussiandistributionswithsu cientlyshorttails.Aresamplingapproachisproposedforchoosingthebandingparameterinpractice.Thisapproachisillustratednumericallyonbothsimulatedandrealdata.
Keywords:covarianceestimation,regularization,bandedestimators,Choleskydecomposition.
1Introduction
Estimationofpopulationcovariancematricesfromsamplesofmultivariatedatahasalwaysbeenimportantforanumberofreasons.Principalamongtheseare:(1)estimationofprincipalcompo-nentsandeigenvaluesinordertogetaninterpretablelow-dimensionaldatarepresentation(principalcomponentanalysis,orPCA);(2)constructionoflineardiscriminantfunctionsforclassi cationofGaussiandata(lineardiscriminantanalysis,orLDA);(3)establishingindependenceandcondi-tionalindependencerelationsbetweencomponentsusingexploratorydataanalysisandtesting;and(4)settingcon denceintervalsonlinearfunctionsofthemeansofthecomponents.Notethat(1)requiresestimationoftheeigenstructureofthecovariancematrixwhile(2)and(3)requireestimationoftheinverse.
Thetheoryofmultivariateanalysisfornormalvariableshasbeenwellworkedout–seeAnderson(1958),themajormonograph.However,itbecameapparentthatexactexpressionswerecumber-some,evenforsmalldimensionsandsamplesizes,andthatmultivariatedatawererarelyGaussian.Theremedywasasymptotictheoryforlargesamplesand xedrelativelysmalldimensions.Inrecentyears,datasetsthatdonot tintothisframeworkhavebecomeverycommon–thedataareveryhigh-dimensionalandsamplesizescanbeverysmallrelativetodimension.Examplesincludegeneexpressionarrays,fMRIdata,spectroscopicimaging,numericalweatherforecasting,andmanyothers.
Regularized estimation o... 暂无评价 30页 免费 Regularized estimation o......Eigenvalues of Large Sample Covariance Matrices of Spiked Population Models Jinho...
Regularized estimation o... 暂无评价 30页 免费 Regularized estimation o......Spectrum estimation for large dimensional covariance matrices using random matrix...
Regularized estimation o... 暂无评价 28页 免费 Covariance Matrices for ......In the limit of suf?ciently large sample sizes in which regularization ...
Regularized estimation o... 暂无评价 30页 免费 Regularized estimation o......theorems on the eigenvectors of large dimensional sample covariance matrices ...
Regularized estimation o... 暂无评价 30页 免费 covariance 24页 免费 Eigenvalues...LINEAR SPECTRAL STATISTICS OF LARGE-DIMENSIONAL SAMPLE COVARIANCE MATRICES B ...
The convergence on spectrum of sample covariance matrices for information-plus-noise type data 隐藏>> Ap1Mah.3.ChisUni p.tneev 21,72:8—9 022()1111 Th...
Spectral Density of Sample Covariance Matrices of Colored Noise_专业资料。We ...When the population size is not large and the number of sampling points ...
Covariance and precision matrix estimation for high-dimensional time series_电子...(2008b). Regularized estimation of large covariance matrices. Ann. Statist....
Regularized estimation o... 暂无评价 28页 免费 MINIMAL DEGREE COPRIM... 暂无...FACTORIZATION OF SEPARABLE AND PATTERNED COVARIANCE MATRICES FOR GIBBS SAMPLING...
Covariance selection and estimation via penalised normal likelihood_电子/电路_工程...(2003). Nonparametric estimation of large covariance matrices of longitudinal ...
热门文档