11.04.2014 Views

Linear Algebra Exercises-n-Answers.pdf

Linear Algebra Exercises-n-Answers.pdf

Linear Algebra Exercises-n-Answers.pdf

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

<strong>Answers</strong> to <strong>Exercises</strong> 41<br />

(d) Yes.<br />

{b<br />

( ) ( )<br />

−1 0 0 1 ∣∣<br />

+ c b, c ∈ R}<br />

0 1 0 0<br />

Two.I.2.21 No, it is not closed. In particular, it is not closed under scalar multiplication because it<br />

does not contain the zero polynomial.<br />

Two.I.2.22<br />

(a) Yes, solving the linear system arising from<br />

⎛<br />

r 1<br />

⎝ 1 ⎞ ⎛<br />

0⎠ + r 2<br />

⎝ 0 ⎞ ⎛<br />

0⎠ = ⎝ 2 ⎞<br />

0⎠<br />

0 1 1<br />

gives r 1 = 2 and r 2 = 1.<br />

(b) Yes; the linear system arising from r 1 (x 2 ) + r 2 (2x + x 2 ) + r 3 (x + x 3 ) = x − x 3<br />

2r 2 + r 3 = 1<br />

r 1 + r 2 = 0<br />

r 3 = −1<br />

gives that −1(x 2 ) + 1(2x + x 2 ) − 1(x + x 3 ) = x − x 3 .<br />

(c) No; any combination of the two given matrices has a zero in the upper right.<br />

Two.I.2.23 (a) Yes; it is in that span since 1 · cos 2 x + 1 · sin 2 x = f(x).<br />

(b) No, since r 1 cos 2 x + r 2 sin 2 x = 3 + x 2 has no scalar solutions that work for all x. For instance,<br />

setting x to be 0 and π gives the two equations r 1 · 1 + r 2 · 0 = 3 and r 1 · 1 + r 2 · 0 = 3 + π 2 , which<br />

are not consistent with each other.<br />

(c) No; consider what happens on setting x to be π/2 and 3π/2.<br />

(d) Yes, cos(2x) = 1 · cos 2 (x) − 1 · sin 2 (x).<br />

Two.I.2.24<br />

(a) Yes, for any x, y, z ∈ R this equation<br />

⎛<br />

r 1<br />

⎝ 1 ⎞ ⎛<br />

0⎠ + r 2<br />

⎝ 0 ⎞ ⎛<br />

2⎠ + r 3<br />

⎝ 0 ⎞ ⎛<br />

0⎠ = ⎝ x ⎞<br />

y⎠<br />

0 0 3 z<br />

has the solution r 1 = x, r 2 = y/2, and r 3 = z/3.<br />

(b) Yes, the equation<br />

⎛<br />

r 1<br />

⎝ 2 ⎞ ⎛<br />

0⎠ + r 2<br />

⎝ 1 ⎞ ⎛<br />

1⎠ + r 3<br />

⎝ 0 ⎞ ⎛<br />

0⎠ = ⎝ x ⎞<br />

y⎠<br />

1 0 1 z<br />

gives rise to this<br />

2r 1 + r 2<br />

r 1<br />

r 2<br />

= x<br />

= y<br />

+ r 3 = z<br />

−(1/2)ρ 1+ρ 3<br />

−→<br />

(1/2)ρ 2+ρ 3<br />

−→ r 2 = y<br />

2r 1 + r 2 = x<br />

r 3 = −(1/2)x + (1/2)y + z<br />

so that, given any x, y, and z, we can compute that r 3 = (−1/2)x + (1/2)y + z, r 2 = y, and<br />

r 1 = (1/2)x − (1/2)y.<br />

(c) No. In particular, the vector<br />

⎛ ⎞<br />

0<br />

⎝0⎠<br />

1<br />

cannot be gotten as a linear combination since the two given vectors both have a third component<br />

of zero.<br />

(d) Yes. The equation<br />

⎛<br />

r 1<br />

⎝ 1 ⎞ ⎛<br />

0⎠ + r 2<br />

⎝ 3 ⎞ ⎛<br />

1⎠ + r 3<br />

⎝ −1<br />

⎞ ⎛<br />

0 ⎠ + r 4<br />

⎝ 2 ⎞ ⎛<br />

1⎠ = ⎝ x ⎞<br />

y⎠<br />

1 0 0 5 z<br />

leads to this reduction.<br />

⎛<br />

⎝ 1 3 −1 2 x<br />

⎞<br />

0 1 0 1 y<br />

1 0 0 5 z<br />

⎠ −ρ1+ρ3<br />

−→<br />

3ρ2+ρ3<br />

−→<br />

⎛<br />

⎝ 1 3 −1 2 x<br />

0 1 0 1 y<br />

0 0 1 6 −x + 3y + z<br />

We have infinitely many solutions. We can, for example, set r 4 to be zero and solve for r 3 , r 2 , and<br />

r 1 in terms of x, y, and z by the usual methods of back-substitution.<br />

⎞<br />

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!