Perturbation expansions for invariant subspaces (Q807712): Difference between revisions

From MaRDI portal
RedirectionBot (talk | contribs)
Removed claim: author (P16): Item:Q226558
ReferenceBot (talk | contribs)
Changed an Item
 
(2 intermediate revisions by 2 users not shown)
Property / author
 
Property / author: Ji-guang Sun / rank
 
Normal rank
Property / MaRDI profile type
 
Property / MaRDI profile type: MaRDI publication profile / rank
 
Normal rank
Property / cites work
 
Property / cites work: Q5800951 / rank
 
Normal rank
Property / cites work
 
Property / cites work: On Multiple Eigenvalues of Matrices Depending on Several Parameters / rank
 
Normal rank
Property / cites work
 
Property / cites work: Q5583194 / rank
 
Normal rank
Property / cites work
 
Property / cites work: Q5638711 / rank
 
Normal rank
Property / cites work
 
Property / cites work: Error and Perturbation Bounds for Subspaces Associated with Certain Eigenvalue Problems / rank
 
Normal rank
Property / cites work
 
Property / cites work: Q3754747 / rank
 
Normal rank
Property / cites work
 
Property / cites work: Q3788137 / rank
 
Normal rank
Property / cites work
 
Property / cites work: Q5674306 / rank
 
Normal rank

Latest revision as of 17:04, 21 June 2024

scientific article
Language Label Description Also known as
English
Perturbation expansions for invariant subspaces
scientific article

    Statements

    Perturbation expansions for invariant subspaces (English)
    0 references
    1991
    0 references
    This is an application of the author's method described in [J. Comput. Math. 3, 351-364 (1985; Zbl 0618.15009)]. The present main result is the following expansion theorem. Let A, B, \(X=(X_ 1,X_ 2),\) \(Y=(Y_ 1,Y_ 2)\in C^{n\times n}\), \(X_ 1\in C^{n\times \ell}\), \(X_ 1^ HX_ 1=I_{\ell}\), \(X_ 2^ HX_ 2=I_{n-\ell}\), \(Y_ 1\in C^{n\times \ell}\) be such that \(Y^ TX=I\), \(Y^ TAX=\left[ \begin{matrix} A_{11}\\ 0\end{matrix} \begin{matrix} A_{12}\\ A_{22}\end{matrix} \right]\) with \(A_{11}\in C^{\ell \times \ell}\), \(\sigma (A_{11})\cap \sigma (A_{22})=\emptyset\), \(\chi_ 1\) the column space of \(X_ 1\), \(Y^ TBX=\left[ \begin{matrix} B_{11}\\ B_{21}\end{matrix} \begin{matrix} B_{12}\\ B_{22}\end{matrix} \right]\), and let the operator T be defined by \(TP=PA_{11}-A_{22}P\) with \(P\in C^{(n-\ell)\times \ell}\). Then \(A+tB\) has a unique \(\ell\)-dimensional invariant subspace \(\chi_ 1(t)\) such that \(\chi_ 1(0)=\chi_ 1\) with basis vectors \(x_ 1(t),...,x_{\ell}(t)\) that are analytic functions in a neighborhood N of \(t=0\) and the matrix valued function \(X_ 1(t)=(x_ 1(t),...,x_{\ell}(t))\) has a perturbation expansion \(X_ 1(t)=X_ 1+\sum^{\infty}_{m=1}X_ 2W_ mt^ m\), \(t\in N\), with \(W_ 1=T^{-1}B_{21}\), \(W_ 2=T^{-1}[(B_{22}W_ 1-W_ 1B_{11})- W_ 1A_{12}W_ 1]\), etc. From this is derived a perturbation estimation at \(t=0\) of any order k.
    0 references
    invariant subspace
    0 references
    perturbation expansion
    0 references
    perturbation estimation
    0 references
    0 references

    Identifiers