A related line of recent work on learning sparse models has focused on \stagewise greedy algorithms. Jan 01, 2011 however, the sample covariance matrix is an inappropriate estimator in high dimensional settings. High dimensional data are often most plausibly generated from distributions with complex structure and leptokurtosis in some or all components. The abundance of high dimensional data is one reason for the interest in the problem. In this paper, we propose a maximum likelihood ml approach to covariance estimation, which employs a novel sparsity constraint. Highdimensional data are often most plausibly generated from distributions with. Covariance and precision matrices provide a useful summary of such structure, yet the performance of popular matrix estimators typically hinges upon a subgaussianity assumption. Estimation of large dimensional sparse covariance matrices. High dimensional inverse covariance matrix estimation via. Highdimensional sparse inverse covariance estimation using.
High dimensional low rank and sparse covariance matrix. Covariance estimation for high dimensional data vectors. Highdimensional covariance matrix estimation in approximate. The key requirement on for optimal covariance estimation is that. This paper studies methods for testing and estimating changepoints in the covariance structure of a high dimensional linear time series. Sisnota good estimator of in fact, i for n estimating high dimensional covariance matrices 201 2. Download citation highdimensional covariance estimation penalized regression methods for inducing sparsity in the precision matrix have grown rapidly in.
Estimation of large covariance matrices, particularly in situations where the data dimension p is comparable to or larger than the sample size n, has attracted a lot of attention recently. High dimensional covariance estimation provides accessible and comprehensive coverage of the classical and modern approaches for estimating covariance matrices as well as their applications to the rapidly developing areas lying at the intersection of statistics and machine learning. Highdimensional covariance estimation provides accessible and comprehensive coverage of the classical and modern approaches for estimating covariance. Highdimensional covariance estimation based on gaussian. Highdimensional covariance estimation researchgate.
When the dimension of the covariance matrix is large, the estimation problem. In this paper, we propose a maximum likelihood ml approach to covariance estimation, which employs a. Jul 26, 20 high dimensional covariance estimation provides accessible and comprehensive coverage of the classical and modern approaches for estimating covariance matrices as well as their applications to the rapidly developing areas lying at the intersection of statistics and machine learning. Highdimensional covariance estimation mohsen pourahmadi. Tony cai department of statistics, the wharton school, university of pennsylvania, philadelphia, pa 19104, usa email. Pdf highdimensional covariance matrix estimation in. Robust estimation of high dimensional covariance and precision matrices by marcoavellamedina sloan school of management, massachusetts institute oftechnology, 30 memorial drive, cambridge, massachusetts 02142, u.
Systems in the university of michigan 2011 doctoral committee. With highdimensional data wiley series in probability and statistics kindle edition by pourahmadi, mohsen. Highdimensional covariance estimation can be classified into two main categories, one. Classical methods of estimating the covariance matrices are based on the strict factor models, assuming independent idiosyncratic. High dimensional covariance estimation by minimizing l1penalized logdeterminant divergence pradeep ravikumar, martin wainwright, bin yu, garvesh raskutti abstract. Fast covariance estimation for highdimensional functional data. An overview on the estimation of large covariance and. Covariance estimation for high dimensional data vectors using the sparse matrix transform guangzhi cao and charles a. In our later simulations and applications, the tolerance value is set to be 0. Bouman school of electrical and computer engineering purdue university west lafayette, in 47907 april 29, 2008 1 introduction many problems in statistical pattern recognition and analysis require the classi cation and.
The method, permuted rankpenalized leastsquares prls, is based on a. Another relation can be made to the method by rutimann. Wainwright, garvesh raskutti, and bin yu more by pradeep ravikumar. This implies that the choice to take the sample covariance as the pilot estimator. Methods for estimating sparse and large covariance matrices covariance and correlation matrices play fundamental roles in every aspect of the analysis of multivariate data collected from a variety of fields including business and economics, health care, engineering, and environmental and physical sciences. High dimensional low rank and sparse covariance matrix estimation via convex minimization. Perhaps the most natural candidate for estimating is the empirical sample covariance matrix, but this is known to behave poorly in highdimensional settings. The assumed framework allows for a large class of multivariate linear processes including vector autoregressive moving average varma models of growing dimension and spiked covariance models. Minimax rates of convergence for estimating several classes of structured covariance and precision matrices, including bandable, toeplitz, and sparse covariance matrices as well as sparse precision matrices, are given under the spectral norm loss.
Minimax rates of convergence for estimating several classes of. Most available methods and software cannot smooth covariance matrices of dimension j \times j with j500. Robust estimation of highdimensional covariance and precision. We study high dimensional covariance precision matrix estimation under the assumption that the covariance precision matrix can be decomposed into a lowrank component l and a diagonal component d.
These perform simple forward steps adding parameters greedily, and possibly also backward steps removing parameters greed. Matlab software for disciplined convex programming, version 2. Asymptotic distributions of highdimensional nonparametric inference with distance correlation. Wolf, a wellconditioned estimator for largedimensional covariance matrices, journal of multivariate analysis, volume 88, issue 2, february 2004, pages 365. Fast covariance estimation for highdimensional functional. This paper presents a new method for estimating high dimensional covariance matrices. Even if p software that scale up linearly with the number of observations per function. Focusing on methodology and computation more than on theorems and proofs, this book provides computationally feasible and statistically efficient methods for estimating sparse and large covariance matrices of high dimensional data. Many techniques for detection and estimation rely on accurate estimation of the true covariance.
Xi luo brown university november 7, 2011 abstract this paper introduces a general framework of covariance structures that can be veri. Covariance estimation in high dimensions via kronecker. Regularized estimation of highdimensional covariance matrices. Regularized estimation of high dimensional covariance matrices by yilun chen a dissertation submitted in partial ful llment of the requirements for the degree of doctor of philosophy electrical engineering. In recent years, estimating a high dimensional p pcovariance matrix under small sample size nhas attracted considerable attention. Rp, estimate both its covariance matrix, and its inverse covariance or concentration matrix.
Estimating and testing highdimensional mediation effects in epigenetic studies. Most available methods and software cannot smooth covariance matrices of dimension j 500. Sparse estimation of highdimensional covariance matrices. Testing and estimating changepoints in the covariance matrix. Battey department of mathematics, imperial college london, 545 huxley building. Highdimensional data are often most plausibly generated from distributions with complex structure and leptokurtosis in some or all components. Robust estimation of highdimensional covariance and. Why we need a sparse estimation of a covariance matrix.
Popular regularization methods of directly exploiting sparsity are not directly applicable to many financial problems. Robust covariance estimation for highdimensional compositional data with application to microbial communities analysis nasaads microbial communities analysis is drawing growing attention due to the rapid development of highthroughput sequencing techniques nowadays. High dimensional covariance estimation provides accessible and comprehensive coverage of. For example, in portfolio allocation and risk management, the number of stocks p, which is typically of the same order as the sample size n, can well be in the order of hundreds. High dimensional sparse inverse covariance estimation using greedy methods recent resurgence of greedy methods.
Estimating high dimensional covariance matrices and its. Rejoinder of estimating structured highdimensional covariance and precision matrices. Minimax rates of convergence for estimating several classes of structured covariance and precision matrices, including bandable, toeplitz, sparse, and sparse spiked covariance matrices as well as. High dimensional covariance matrix estimation using a factor. Estimating structured highdimensional covariance and. Estimating high dimensional covariance matrices is intrinsically challenging. For example, x itcan be the return for asset iin period t, t 1. Rp, we study the problem of estimating both its covariance matrix, and its inverse covariance or concentration matrix. Robust shrinkage estimation of highdimensional covariance. Estimating structured high dimensional covariance and precision matrices.
Instead, we introduce robust pilot estimators in 4 that satisfy the conditions 1 and 2. High dimensional covariance matrix estimation ming yuan department of statistics university of wisconsinmadison and morgridge institute for research. The limitations of the sample covariance matrix are discussed. When n high dimensional covariance estimation provides accessible and comprehensive coverage of the classical and modern approaches for estimating covariance matrices as well as their applications to the rapidly developing areas lying at the intersection of statistics and machine learning. High dimensional covariance estimation by minimizing l1. Highdimensional covariance estimation provides accessible and comprehensive coverage of the classical and modern approaches for estimating covariance matrices as well as their applications to the rapidly developing areas lying at the intersection of statistics and machine learning. Covariance estimation for high dimensional vectors is a classically dif. High dimensional inverse covariance matrix estimation via linear programming. May 21, 2011 the variance covariance matrix plays a central role in the inferential theories of high dimensional factor models in finance and economics.
In the following, nis referred to as the number of variables, or the number. Estimating covariance matrices is an important part of portfolio selection, risk management, and asset pricing. Download it once and read it on your kindle device, pc, phones or tablets. Highdimensional covariance estimation by minimizing. High dimensional covariance matrix estimation by penalizing.