Second-order accurate inference on eigenvalues of covariance and correlation matrices

Edgeworth expansions and saddlepoint approximations for the distributions of estimators of certain eigenfunctions of covariance and correlation matrices are developed. These expansions depend on second-, third-, and fourth-order moments of the sample covariance matrix. Expressions for and estimators of these moments are obtained. The expansions and moment expressions are used to construct second-order accurate confidence intervals for the eigenfunctions. The expansions are illustrated and the results of a small simulation study that evaluates the finite-sample performance of the confidence intervals are reported.     

Maximum likelihood factor analysis with rank-deficient sample covariance matrices

This paper characterises completely the circumstances in which maximum likelihood estimation of the factor model is feasible when the sample covariance matrix is rank deficient. This situation will arise when the number of variables exceeds the number of observations.     

A well-conditioned estimator for large-dimensional covariance matrices

Many applied problems require a covariance matrix estimator that is not only invertible, but also well-conditioned (that is, inverting it does not amplify estimation error). For large-dimensional covariance matrices, the usual estimator—the sample covariance matrix—is typically not well-conditioned and may not even be invertible. This paper introduces an estimator that is both well-conditioned and more accurate than the sample covariance matrix asymptotically. This estimator is distribution-free and has a simple explicit formula that is easy to compute and interpret. It is the asymptotically optimal convex linear combination of the sample covariance matrix with the identity matrix. Optimality is meant with respect to a quadratic loss function, asymptotically as the number of observations and the number of variables go to infinity together. Extensive Monte Carlo confirm that the asymptotic results tend to hold well in finite sample.     

Canonical spectral equation for empirical covariance matrices

We study asymptotic properties of normalized spectral functions of empirical covariance matrices in the case of a nonnormal population. It is shown that the Stieltjes transforms of such functions satisfy a socalled canonical spectral equation.Translated from Ukrainskii Matematicheskii Zhurnal, Vol. 47, No. 9, pp. 1176–1189, September, 1995.     

Proportionality of k covariance matrices

This article discusses maximum likelihood estimation of proportional covariance matrices under normality assumptions. An algorithm for solving the likelihood equations and the likelihood ratio statistic for testing the hypothesis of proportionality are given. The method is illustrated by a numerical example.     

Partial estimation of covariance matrices

A classical approach to accurately estimating the covariance matrix Σ of a p-variate normal distribution is to draw a sample of size n > p and form a sample covariance matrix. However, many modern applications operate with much smaller sample sizes, thus calling for estimation guarantees in the regime ${n \ll p}$ . We show that a sample of size n = O(m log⁶ p) is sufficient to accurately estimate in operator norm an arbitrary symmetric part of Σ consisting of m ≤ n nonzero entries per row. This follows from a general result on estimating Hadamard products M · Σ, where M is an arbitrary symmetric matrix.     

Sensitivity analysis and calibration of the covariance matrix for stable portfolio selection

We recommend an implementation of the Markowitz problem to generate stable portfolios with respect to perturbations of the problem parameters. The stability is obtained proposing novel calibrations of the covariance matrix between the returns that can be cast as convex or quasiconvex optimization problems. A statistical study as well as a sensitivity analysis of the Markowitz problem allow us to justify these calibrations. Our approach can be used to do a global and explicit sensitivity analysis of a class of quadratic optimization problems. Numerical simulations finally show the benefits of the proposed calibrations using real data.     

On testing for homogeneity of the covariance matrices

Testing equality of covariance matrix ofk populations has long been an interesting issue in statistical inference. To overcome the sparseness of data points in a high-dimensional space and deal with the general cases, we suggest several projection pursuit type statistics. Some results on the limiting distributions of the statistics are obtained. Some properties of Bootstrap approximation are investigated. Furthermore, for computational reasons an approximation which is based on Number theoretic method for the statistics is adopted. Several simulation experiments are performed.     

Two step-down tests for equality of covariance matrices

The classical problem of testing the equality of the covariance matrices from k ? 2 p-dimensional normal populations is reexamined. The likelihood ratio (LR) statistic, also called Bartlett’s statistic, can be decomposed in two ways, corresponding to two distinct component-wise decompositions of the null hypothesis in terms of the covariance matrices or precision matrices, respectively. The factors of the LR statistic that appear in these two decompositions can be interpreted as conditional and unconditional LR statistics for the component-wise null hypotheses, and their mutual independence under the null hypothesis allows the determination of the overall significance level.     

Butson Hadamard matrices with partially cyclic core

In this paper, we introduce a class of generalized Hadamard matrices, called a Butson Hadamard matrix with partially cyclic core. Then a new construction method for Butson Hadamard matrices with partially cyclic core is proposed. The proposed matrices are constructed from the optimal balanced low-correlation zone(LCZ) sequence set which has correlation value ?1 within LCZ.     

Randomly generating portfolio-selection covariance matrices with specified distributional characteristics

In portfolio selection, there is often the need for procedures to generate “realistic” covariance matrices for security returns, for example to test and benchmark optimization algorithms. For application in portfolio optimization, such a procedure should allow the entries in the matrices to have distributional characteristics which we would consider “realistic” for security returns. Deriving motivation from the fact that a covariance matrix can be viewed as stemming from a matrix of factor loadings, a procedure is developed for the random generation of covariance matrices (a) whose off-diagonal (covariance) entries possess a pre-specified expected value and standard deviation and (b) whose main diagonal (variance) entries possess a likely different pre-specified expected value and standard deviation. The paper concludes with a discussion about the futility one would likely encounter if one simply tried to invent a valid covariance matrix in the absence of a procedure such as in this paper.     

Modeling covariance matrices via partial autocorrelations

We study the role of partial autocorrelations in the reparameterization and parsimonious modeling of a covariance matrix. The work is motivated by and tries to mimic the phenomenal success of the partial autocorrelations function (PACF) in model formulation, removing the positive-definiteness constraint on the autocorrelation function of a stationary time series and in reparameterizing the stationarity-invertibility domain of ARMA models. It turns out that once an order is fixed among the variables of a general random vector, then the above properties continue to hold and follow from establishing a one-to-one correspondence between a correlation matrix and its associated matrix of partial autocorrelations. Connections between the latter and the parameters of the modified Cholesky decomposition of a covariance matrix are discussed. Graphical tools similar to partial correlograms for model formulation and various priors based on the partial autocorrelations are proposed. We develop frequentist/Bayesian procedures for modelling correlation matrices, illustrate them using a real dataset, and explore their properties via simulations.     

Spectra of infinite-dimensional sample covariance matrices

We study spectral functions of infinite-dimensional random Gram matrices of the form RR^T, where R is a rectangular matrix with an infinite number of rows and with the number of columns N → ∞, and the spectral functions of infinite sample covariance matrices calculated for samples of volume N → ∞ under conditions analogous to the Kolmogorov asymptotic conditions. We assume that the traces d of the expectations of these matrices increase with the number N such that the ratio d/N tends to a constant. We find the limiting nonlinear equations relating the spectral functions of random and nonrandom matrices and establish the asymptotic expression for the resolvent of random matrices. __________ Translated from Teoreticheskaya i Matematicheskaya Fizika, Vol. 148, No. 2, pp. 309–322, August, 2006.     

Optimal rates of convergence for estimating Toeplitz covariance matrices

Toeplitz covariance matrices are used in the analysis of stationary stochastic processes and a wide range of applications including radar imaging, target detection, speech recognition, and communications systems. In this paper, we consider optimal estimation of large Toeplitz covariance matrices and establish the minimax rate of convergence for two commonly used parameter spaces under the spectral norm. The properties of the tapering and banding estimators are studied in detail and are used to obtain the minimax upper bound. The results also reveal a fundamental difference between the tapering and banding estimators over certain parameter spaces. The minimax lower bound is derived through a novel construction of a more informative experiment for which the minimax lower bound is obtained through an equivalent Gaussian scale model and through a careful selection of a finite collection of least favorable parameters. In addition, optimal rate of convergence for estimating the inverse of a Toeplitz covariance matrix is also established.     

Efficient statistical inference for partially nonlinear errors-in-variables models

In this paper, we consider the partially nonlinear errors-in-variables models when the non- parametric component is measured with additive error. The profile nonlinear least squares estimator of unknown parameter and the estimator of nonparametric component are constructed, and their asymptotic properties are derived under general assumptions. Finite sample performances of the proposed statistical inference procedures are illustrated by Monte Carlo simulation studies.