Explore the vast range of titles available.

Estimating covariance matrices is a problem of fundamental importance in multivariate statistics. In practice it is increasingly frequent to work with data matrices X of dimension n × p, where p and n are both large. Results from random matrix theory show very clearly that in this setting, standard estimators like the sample covariance matrix perform in general very poorly. In this \"large n, large p\" setting, it is sometimes the case that practitioners are willing to assume that many elements of the population covariance matrix are equal to 0, and hence this matrix is sparse. We develop an estimator to handle this situation. The estimator is shown to be consistent in operator norm, when, for instance, we have $p\\asymp n$ as n → ∞. In other words the largest singular value of the difference between the estimator and the population covariance matrix goes to zero. This implies consistency of all the eigenvalues and consistency of eigenspaces associated to isolated eigenvalues. We also propose a notion of sparsity for matrices, that is, \"compatible\" with spectral analysis and is independent of the ordering of the variables.

The portfolio optimization model has limited impact in practice because of estimation issues when applied to real data. To address this, we adapt two machine learning methods, regularization and cross-validation, for portfolio optimization. First, we introduce performance-based regularization (PBR), where the idea is to constrain the sample variances of the estimated portfolio risk and return, which steers the solution toward one associated with less estimation error in the performance. We consider PBR for both mean-variance and mean-conditional value-at-risk (CVaR) problems. For the mean-variance problem, PBR introduces a quartic polynomial constraint, for which we make two convex approximations: one based on rank-1 approximation and another based on a convex quadratic approximation. The rank-1 approximation PBR adds a bias to the optimal allocation, and the convex quadratic approximation PBR shrinks the sample covariance matrix. For the mean-CVaR problem, the PBR model is a combinatorial optimization problem, but we prove its convex relaxation, a quadratically constrained quadratic program, is essentially tight. We show that the PBR models can be cast as robust optimization problems with novel uncertainty sets and establish asymptotic optimality of both sample average approximation (SAA) and PBR solutions and the corresponding efficient frontiers. To calibrate the right-hand sides of the PBR constraints, we develop new, performance-based k -fold cross-validation algorithms. Using these algorithms, we carry out an extensive empirical investigation of PBR against SAA, as well as L1 and L2 regularizations and the equally weighted portfolio. We find that PBR dominates all other benchmarks for two out of three Fama–French data sets. This paper was accepted by Yinyu Ye, optimization .

Estimating the eigenvalues of a population covariance matrix from a sample covariance matrix is a problem of fundamental importance in multivariate statistics; the eigenvalues of covariance matrices play a key role in many widely used techniques, in particular in principal component analysis (PCA). In many modern data analysis problems, statisticians are faced with large datasets where the sample size, n, is of the same order of magnitude as the number of variables p. Random matrix theory predicts that in this context, the eigenvalues of the sample covariance matrix are not good estimators of the eigenvalues of the population covariance. We propose to use a fundamental result in random matrix theory, the Marčenko-Pastur equation, to better estimate the eigenvalues of large dimensional covariance matrices. The Marčenko-Pastur equation holds in very wide generality and under weak assumptions. The estimator we obtain can be thought of as \"shrinking\" in a nonlinear fashion the eigenvalues of the sample covariance matrix to estimate the population eigenvalues. Inspired by ideas of random matrix theory, we also suggest a change of point of view when thinking about estimation of high-dimensional vectors: we do not try to estimate directly the vectors but rather a probability measure that describes them. We think this is a theoretically more fruitful way to think about these problems. Our estimator gives fast and good or very good results in extended simulations. Our algorithmic approach is based on convex optimization. We also show that the proposed estimator is consistent.

We study ridge-regularized generalized robust regression estimators, i.e. β^=argminβ∈Rp1n∑i=1nρi(Yi-Xi′β)+τ2||β||2,whereYi=ϵi+Xi′β0,in the situation where p/n tends to a finite non-zero limit. Our study here focuses on the situation where the errors ϵi’s are heavy-tailed and Xi’s have an “elliptical-like” distribution. Our assumptions are quite general and we do not require homoskedasticity of ϵi’s for instance. We obtain a characterization of the limit of ||β^-β0||, as well as several other results, including central limit theorems for the entries of β^.

We consider the asymptotic fluctuation behavior of the largest eigenvalue of certain sample covariance matrices in the asymptotic regime where both dimensions of the corresponding data matrix go to infinity. More precisely, let X be an n × p matrix, and let its rows be i.i.d. complex normal vectors with mean 0 and covariance $\\Sigma _{p}$. We show that for a large class of covariance matrices $\\Sigma _{p}$, the largest eigenvalue of X*X is asymptotically distributed (after recentering and rescaling) as the Tracy-Widom distribution that appears in the study of the Gaussian unitary ensemble. We give explicit formulas for the centering and scaling sequences that are easy to implement and involve only the spectral distribution of the population covariance, n and p. The main theorem applies to a number of covariance models found in applications. For example, well-behaved Toeplitz matrices as well as covariance matrices whose spectral distribution is a sum of atoms (under some conditions on the mass of the atoms) are among the models the theorem can handle. Generalizations of the theorem to certain spiked versions of our models and a.s. results about the largest eigenvalue are given. We also discuss a simple corollary that does not require normality of the entries of the data matrix and some consequences for applications in multivariate statistics.

We place ourselves in the setting of high-dimensional statistical inference where the number of variables p in a dataset of interest is of the same order of magnitude as the number of observations n. We consider the spectrum of certain kernel random matrices, in particular n × n matrices whose (i, j)th entry is $f(X_{i}^{\\prime }X_{j}/p)$ or f(∥X i – X j ∥²/p) where p is the dimension of the data, and X i are independent data vectors. Here f is assumed to be a locally smooth function. The study is motivated by questions arising in statistics and computer science where these matrices are used to perform, among other things, nonlinear versions of principal component analysis. Surprisingly, we show that in high-dimensions, and for the models we analyze, the problem becomes essentially linear—which is at odds with heuristics sometimes used to justify the usage of these methods. The analysis also highlights certain peculiarities of models widely studied in random matrix theory and raises some questions about their relevance as tools to model high-dimensional data encountered in practice.

We first study the properties of solutions of quadratic programs with linear equality constraints whose parameters are estimated from data in the high-dimensional setting where p, the number of variables in the problem, is of the same order of magnitude as n, the number of observations used to estimate the parameters. The Markowitz problem in Finance is a subcase of our study. Assuming normality and independence of the observations we relate the efficient frontier computed empirically to the \"true\" efficient frontier. Our computations show that there is a separation of the errors induced by estimating the mean of the observations and estimating the covariance matrix. In particular, the price paid for estimating the covariance matrix is an underestimation of the variance by a factor roughly equal to 1 - p/n. Therefore the risk of the optimal population solution is underestimated when we estimate it by solving a similar quadratic program with estimated parameters. We also characterize the statistical behavior of linear functionals of the empirical optimal vector and show that they are biased estimators of the corresponding population quantities. We investigate the robustness of our Gaussian results by extending the study to certain elliptical models and models where our n observations are correlated (in \"time\"). We show a lack of robustness of the Gaussian results, but are still able to get results concerning first order properties of the quantities of interest, even in the case of relatively heavy-tailed data (we require two moments). Risk underestimation is still present in the elliptical case and more pronounced than in the Gaussian case. We discuss properties of the nonparametric and parametric bootstrap in this context. We show several results, including the interesting fact that standard applications of the bootstrap generally yield inconsistent estimates of bias. We propose some strategies to correct these problems and practically validate them in some simulations. Throughout this paper, we will assume that p, n and n — p tend to infinity, and p

Journal Article

Share this book

Add to My Shelf