COMPUTATIONAL METHODS OF LINEAR ALGEBRA

The numerical solution of the problems of mathematical physics is most frequently connected with the numerical solution of basic problems of linear algebra—that of solving a system of linear equations, and that of the computation of the proper numbers of a matrix. The present book is an endeavor at systematizing the most important numerical methods of linear algebra—classical ones and those developed quite recently as well.

The author docs not pretend to an exhaustive completeness, having included an exposition of those methods only that have already been tested in practice. In the exposition the author has not strived for an irreproachable rigor, and has not analysed all conceivable cases and sub-cases arising in the application of this or that method, having limited herself to the most typical and practical important cases.

The book consists of three chapters. In the first chapter is given the material from linear algebra that is indispensable to what follows. The second chapter is devoted to the numerical solution of systems of linear equations and parallel questions. Lastly, the third chapter contains a description of numerical methods of computing the proper numbers and proper vectors of a matrix.