Sufficient ensemble size for random matrix theory-based handling of singular covariance matrices

Singular covariance matrices are frequently encountered in both machine learning and optimization problems, most commonly due to high dimensionality of data and insufficient sample sizes. Among many methods of regularization, here we focus on a relatively recent random matrix-theoretic approach, the idea of which is to create well-conditioned approximations of a singular covariance matrix and its inverse by taking the expectation of its random projections. We are interested in the error of a Monte Carlo implementation of this approach, which allows subsequent parallel processing in low dimensions in practice. We find that [Formula: see text] random projections, where [Formula: see text] is the size of the original matrix, are sufficient for the Monte Carlo error to become negligible, in the sense of expected spectral norm difference, for both covariance and inverse covariance approximation, in the latter case under mild assumptions.

An Iterative Approach to Ill-Conditioned Optimal Portfolio Selection

Covariance matrix of the asset returns plays an important role in the portfolio selection. A number of papers is focused on the case when the covariance matrix is positive definite. In this paper, we consider portfolio selection with a singular covariance matrix. We describe an iterative method based on a second order damped dynamical systems that solves the linear rank-deficient problem approximately. Since the solution is not unique, we suggest one numerical solution that can be chosen from the iterates that balances the size of portfolio and the risk. The numerical study confirms that the method has good convergence properties and gives a solution as good as or better than the solutions that are based on constrained least norm Moore–Penrose, Lasso, and naive equal-weighted approaches. Finally, we complement our result with an empirical study where we analyze a portfolio with actual returns listed in S&P 500 index.

Portfolio Optimization of the Mean-Absolute Deviation Model of Some Stocks using the Singular Covariance Matrix

Investing in the stock sector, investors often face risk problems. Usually, forming an investment portfolio is done to minimize risk. In this research, investment portfolio optimization is discussed. The data analyzed are 8 shares traded on the capital market in Indonesia through the Indonesia Stock Exchange (IDX). Optimization is performed using the Mean-Absolute Deviation model with the singular covariance matrix to determine the optimal weights. The results of portfolio optimization Mean-Absolute Deviation model with singular covariance matrix method, was obtained optimal portfolio weights that is of 17.22% for BBCA shares; 26.64% for TKIM shares; 9.96% for BBRI shares; 9.96% for BBNI shares; 8.70% for BMRI shares; 3.75% for ADRO shares; 6.52% for GGRM shares; and 17.25% for UNTR shares. Where the optimal portfolio composition is obtained the expected rate of return (expected return) of 0.18% with a portfolio risk level (standard deviation) of 0.07%.

Computing the Moments of the Complex Gaussian: Full and Sparse Covariance Matrix

Given a multivariate complex centered Gaussian vector Z = ( Z 1 , ⋯ , Z p ) with non-singular covariance matrix Σ , we derive sufficient conditions on the nullity of the complex moments and we give a closed-form expression for the non-null complex moments. We present conditions for the factorisation of the complex moments. Computational consequences of these results are discussed.