Amazoncomexecobidosasin0898714540 and by using this

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: } must be a Jordan chain such that (A − I)x1 = 0 and (A − I)x2 = x1 , while x3 is another eigenvector (not dependent on x1 ). Suppose we try to build the Jordan chains in P by starting with the eigenvectors ⎛ ⎞ ⎛⎞ 1 0 x1 = ⎝ 0 ⎠ and x3 = ⎝ 1 ⎠ (7.8.7) −2 0 D E T H IG R obtained by solving (A − I)x = 0 with straightforward Gauss–Jordan elimination. This naive approach fails because x1 ∈ R (A − I) means (A − I)x2 = x1 is an inconsistent system, so x2 cannot be determined. Similarly, x3 ∈ R (A − I) insures that the same difficulty occurs if x3 is used in place of x1 . In other words, even though the vectors in (7.8.7) constitute an otherwise perfectly good basis for N (A − I), they are not suitable for building Jordan chains. You are asked in Exercise 7.8.2 to find the correct basis for N (A − I) that will yield the Jordan chains that constitute P. Y P Example 7.8.4 O C Problem: What do the results concerning the Jordan form for A ∈ C n×n say about the decomposition of C n into invariant subspaces? Solution: Consider P−1 AP = J = diag (J(λ1 ), J(λ2 ), . . . , J(λs )) , where the J(λj ) ’s are the Jordan segments and P = P1 | P2 | · · · | Ps is a matrix of Jordan chains as described in (7.8.5) and on p. 594. If A is considered as a linear operator on C n , and if the set of columns in Pi is denoted by Ji , then the results in §4.9 (p. 259) concerning invariant subspaces together with those in §5.9 (p. 383) about direct sum decompositions guarantee that each R (Pi ) is an invariant subspace for A such that C n = R (P1 ) ⊕ R (P2 ) ⊕ · · · ⊕ R (Ps ) and J(λi ) = A/ R(Pi ) Ji . More can be said. If alg mult (λi ) = mi and index (λi ) = ki , then Ji is a linearly independent set containing mi vectors, and the discussion surrounding Copyright c 2000 SIAM Buy online from SIAM http://www.ec-securehost.com/SIAM/ot71.html Buy from AMAZON.com 596 Chapter 7 Eigenvalues and Eigenvectors http://www.amazon.com/exec/obidos/ASIN/0898714540 (7.8.5) insures that each column in Ji belongs to N (A...
View Full Document

Ask a homework question - tutors are online