A new error in variables model for solving positive definite linear system using orthogonal matrix decompositions

DOI10.1007/S11075-015-0042-2zbMATH Open1355.65046arXiv1407.1372OpenAlexW1932516518MaRDI QIDQ285053FDOQ285053

Authors: Negin Bagherpour, Nezam Mahdavi-Amiri 

Publication date: 18 May 2016

Published in: Numerical Algorithms (Search for Journal in Brave)

Abstract: The need to estimate a positive definite solution to an overdetermined linear system of equations with multiple right hand side vectors arises in several process control contexts. The coefficient and the right hand side matrices are respectively named data and target matrices. A number of optimization methods were proposed for solving such problems, in which the data matrix is unrealistically assumed to be error free. Here, considering error in measured data and target matrices, we present an approach to solve a positive definite constrained linear system of equations based on the use of a newly defined error function. To minimize the defined error function, we derive necessary and sufficient optimality conditions and outline a direct algorithm to compute the solution. We provide a comparison of our proposed approach and two existing methods, the interior point method and a method based on quadratic programming. Two important characteristics of our proposed method as compared to the existing methods are computing the solution directly and considering error both in data and target matrices. Moreover, numerical test results show that the new approach leads to smaller standard deviations of error entries and smaller effective rank as desired by control problems. Furthermore, in a comparative study, using the Dolan-Mor'{e} performance profiles, we show the approach to be more efficient.

Full work available at URL: https://arxiv.org/abs/1407.1372

Recommendations

• A competitive error in variables approach and algorithms for finding positive definite solutions of linear systems of matrix equations
• scientific article; zbMATH DE number 4043148
• scientific article; zbMATH DE number 1932410
• scientific article; zbMATH DE number 4087843
• Criterion of positive definiteness of matrices and solution of inverse problem for system of linear equations

zbMATH Keywords

quadratic programmingalgorithmcomplexityorthogonal decompositioninterior point methodnumerical experimentspectral decompositionpositive definite solutionCholesky decompositonerror in variables modelsmultiple right hand side vectorsoverdetermined linear system of equationspositive definiteness constraints

Mathematics Subject Classification ID

Numerical mathematical programming methods (65K05) Quadratic programming (90C20) Direct numerical methods for linear systems and matrix inversion (65F05) Numerical solutions to overdetermined systems, pseudoinverses (65F20) Interior-point methods (90C51)

Cites Work

• Title not available (Why is that?)
• Functions of Matrices
• Numerical Optimization
• Title not available (Why is that?)
• Title not available (Why is that?)
• Benchmarking optimization software with performance profiles.
• Title not available (Why is that?)
• An Analysis of the Total Least Squares Problem
• Title not available (Why is that?)
• Primal-Dual Interior-Point Methods for Semidefinite Programming: Convergence Rates, Stability and Numerical Results
• GMRES methods for least squares problems
• Title not available (Why is that?)
• Editorial: Total least squares and errors-in-variables modeling
• Perturbation analysis and condition numbers of scaled total least squares problems
• Algebraic connections between the least squares and total least squares problems
• A numerical procedure for finding the positive definite matrix closest to a patterned matrix
• Scaled total least squares fundamentals
• Positive definite constrained least-squares estimation of matrices
• Least-squares solution of \(F=PG\) over positive semidefinite symmetric \(P\)
• Trace inequalities and quantum entropy: an introductory course
• The total least squares problem in \(AX \approx B\): A new classification with the relationship to the classical works
• Maximum Likelihood Estimation of a Structured Covariance Matrix With a Condition Number Constraint
• Benchmarking Derivative-Free Optimization Algorithms
• Least Squares Estimation of the Components of a Symmetric Matrix
• Spectral norm of products of random and deterministic matrices

Cited In (1)

• A competitive error in variables approach and algorithms for finding positive definite solutions of linear systems of matrix equations

Uses Software

• VanHuffel
• mftoolbox

This page was built for publication: A new error in variables model for solving positive definite linear system using orthogonal matrix decompositions

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q285053)