Introduction to Linear algebra-Strang-Solutions-Manual_ver13

# I i 2 i 3 i 4 34 detp 35 3 by 3 permutation

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

This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: e P 2 D P leads to 2 D  so  D 0 or 1. 13 (a) P u D .uuT /u D u.uT u/ D u so  D 1 (b) P v D .uuT /v D u.uT v/ D 0 (c) x 1 D . 1; 1; 0; 0/, x 2 D . 3; 0; 1; 0/, x 3 D . 5; 0; 0; 1/ all have P x D 0x D 0. 14 Two eigenvectors of this rotation matrix are x 1 D .1; i / and x 2 D .1; i / (more generally c x 1 , and d x 2 with cd ¤ 0). 15 The other two eigenvalues are  D 1 . 2 16 Set  D 0 in det.A p 1 ˙ i 3/; the three eigenvalues are 1; 1; 1. I / D .1 / : : : .n / to ﬁnd det A D .1 /.2 /    .n /. p p 17 1 D C d C .a d /2 C 4bc/ and 2 D 1 .a C d / add to a C d . 2 2 If A has 1 D 3 and 2 D 4 then det.A I / D . 3/. 4/ D  7 C 12.     40 32 22 18 These 3 matrices have  D 4 and 5, trace 9, det 20: ; ; . 05 16 37 1 .a 2 19 (a) rank D 2 (d) eigenvalues of .B 2 C I / 1 1 are 1; 2 ; 1 . 5  01 has trace 11 and determinant 28, so  D 4 and 7. Moving to a 3 by 3 28 11 &quot; # 0 10 0 1 has det.C I / D 3 C 62 11 C 6 D companion matrix, C D 0 6 11 6 .1 /.2 /.3 /. Notice the trace 6 D 1 C 2 C 3, determinant 6 D .1/.2/.3/, and also 11 D .1/.2/ C .1/.3/ C .2/.3/. 20 A D  (b) det.B T B/ D 0 Solutions to Exercises 60     I / has the same determinant as .A I /T 10 1 1 have different and 0 0 eigenvectors. because every square matrix has det M D det M T . 1 0 21 .A 22  D 1 (for Markov), 0 (for singular), 23  0 1 1 2 (so sum of eigenvalues D trace D 1 /. 2    0 01 11 Always A2 is the zero matrix if  D 0 and 0, , , . 0 00 11 by the Cayley-Hamilton Theorem in Problem 6.2.32. 24  D 0; 0; 6 (notice rank 1 and trace 6) with x 1 D .0; 2 ; 1/, x 2 D .1; 2 ; 0/, x 3 D .1; 2; 1/. 25 With the same n ’s and x ’s, Ax D c1 1 x 1 C    C cn n x n equals B x D c1 1 x 1 C    C cn n x n for all vectors x . So A D B . 26 The block matrix has  D 1, 2 from B and 5, 7 from D . All entries of C are multiplied by zeros in det.A I /, so C has no effect on the eigenvalues. 27 A has rank 1 with eigenvalues 0; 0; 0; 4 (the 4 comes from the trace of A). C has rank 2 (ensuring two zero eigenvalues) and .1; 1; 1; 1/ is an eigenvector with  D 2. With trace 4, the other eigenvalue is also  D 2, and its eigenvector is .1; 1; 1; 1/. 28 B has  D 1, 1, 1, 3 and C has  D 1; 1; 1; 3. Both have det D 3. p p 3; Rank-1 matrix: .C / D 29 Triangular matrix: .A/ D 1; 4; 6; .B/ D 2, 3, 0; 0; 6.       ab 1 aCb 1 30 D D .a C b / ; 2 D d b to produce the correct trace cd 1 cCd 1 .a C b/ C .d b / D a C d . 31 Eigenvector .1; 3; 4/ for A with  D 11 and eigenvector .3; 1; 4/ for PAP T . Eigenvec- tors with  ¤ 0 must be in the column space since Ax is always in the column space, and x D Ax =. 32 (a) u is a basis for the nullspace, v and w give a basis for the column space (b) x D .0; 1 ; 1 / is a particular solution. Add any c u from the nullspace 35 (c) If Ax D u had a solution, u would be in the column space: wrong dimension 3. 33 If vT u D 0 then A2 D u.vT u/vT is the zero matrix and 2 D 0; 0 and  D 0; 0 and trace .A/ D 0. This z...
View Full Document

Ask a homework question - tutors are online