×

Quotient-difference type generalizations of the power method and their analysis. (English) Zbl 0714.65060

The paper deals with the spectral analysis of vector sequences x(m) (m\(\geq 0)\) for which \(x(m)\sim \sum \nu (j)\lambda (j)^ m\) (0\(\leq j\leq \infty)\) for large m, where the \(\nu\) (j) are linearly independent vectors in a normed linear space \({\mathfrak B}\) and the \(\lambda\) (j) are scalars for which \(| \lambda (1)| \geq | \lambda (2)| \geq....\) Use is made of determinantal expressions G, F and T defined in terms of a double sequence \(\mu\) and a simple sequence b. With m,n\(\geq 0\), \(k\geq 1\), G(\(\mu|\) m,n;k) is the determinant formed from \(\mu (m+i-1,n+j-1)\) (1\(\leq i,j\leq k)\), F(b,\(\mu| m,n;k)\) is formed from the row \(b(n+j-1)\) \((1\leq j\leq k+1)\) followed by the first k rows of \(G(\mu | m,n;k+1)\), and T(b,\(\mu| m,n;k)\) is the quotient \(F(b,\mu | m,n;k)/G(\mu | m,n+1;k)\). Q(1),Q(2),... being linearly independent bounded linear functionals on \({\mathfrak B}\), \(\mu\) can be defined by setting \(\mu (m,n)=Q(m+1| \quad x(n)).\)
It is found that quotients T of contiguous orders satisfy three term recurrence relationships containing a coefficient d(n,k) which, subject to suitable conditions, tends to \(\lambda\) (k) as n increases. Again, a single functional Q may be used to define \(\mu (m,n)=Q(x(m+n))\) and a similar result holds. Further processes of the same type are examined.
Reviewer: P.Wynn

MSC:

65T50 Numerical methods for discrete and fast Fourier transforms
65F15 Numerical computation of eigenvalues and eigenvectors of matrices
47A75 Eigenvalue problems for linear operators
Full Text: DOI

References:

[1] Brezinski, C., Computation of the eigenelements of a matrix by the ϵ-algorithm, Linear Algebra Appl., 11, 7-20 (1975) · Zbl 0311.65028
[2] Ford, W. F.; Sidi, A., Recursive algorithms for vector extrapolation methods, Appl. Numer. Math., 4, 6, 477-489 (1988) · Zbl 0649.65004
[3] Henrici, P., Applied and Computational Complex Analysis, 1 (1974), Wiley: Wiley New York · Zbl 0313.30001
[4] Householder, A. S., The Numerical Treatment of a Single Nonlinear Equation (1970), McGraw-Hill: McGraw-Hill New York · Zbl 0242.65047
[5] Parlett, B. N., The LU and QR algorithms, (Ralston, A.; Wilf, H. S., Mathematical Methods for Digital Computers, II (1967), Wiley: Wiley New York), 116-130 · Zbl 0183.18601
[6] Sidi, A., Convergence and stability properties of minimal polynomial and reduced rank extrapolation algorithms, SIAM J. Numer. Anal., 23, 197-209 (1986) · Zbl 0612.65001
[7] Sidi, A., On extensions of the power method for normal operators, Linear Algebra Appl., 120, 207-224 (1989) · Zbl 0678.65023
[8] Sidi, A.; Ford, W. F.; Smith, D. A., Acceleration of convergence of vector sequences, SIAM J. Numer. Anal., 23, 178-196 (1986) · Zbl 0596.65016
[9] Wynn, P., On a device for computing the \(e_m\)(\(S_n\)) transformation, MTAC, 10, 91-96 (1956) · Zbl 0074.04601
[10] Wynn, P., Acceleration techniques for iterated vector and matrix problems, Math. Comp., 16, 301-322 (1962) · Zbl 0105.10302
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.