Math 115a: Selected Solutions For HW 7: Paul Young November 19, 2005
Uploaded by
Qaty's Morelos RamosMath 115a: Selected Solutions For HW 7: Paul Young November 19, 2005
Uploaded by
Qaty's Morelos RamosMath 115a: Selected Solutions for HW 7
Paul Young
November 19, 2005
Exercise 5.2.2b: For each of the following matrices A Mn (R), test A for
diagonalizability, and if A is diagonalizable, find an invertible matrix Q and a
diagonal matrix D such that Q1 AQ = D.
1 3
A=
3 1
Solution: The characteristic polynomial is
p(t) = (1 t)2 9
= t2 2t 8
= (t 4)(t + 2).
Since the dimension of our vector space is 2 and we have found 2 distinct eigen-
values: 4 and -2, we conclude that A is diagonalizable. Furthermore,
1 1
Q= .
1 1
Exercise 5.2.3b: For each of the following linear operators T on a vector
space V , test T for diagonalizability, and if T is diagonalizable, find a basis
for V such that [T ] is a diagonal matrix: V = P2 (R) and T is defined by
T (ax2 + bx + c) = cx2 + bx + a.
Solution: By inspection, we see that T (x2 + x + 1) = x2 + x + 1 and T (x) = x.
Therefore Eig1 (T ) = span({x2 + x + 1, x}). Secondly, we see that T (x2 1) =
(x2 1) and therefore Eig1 (T ) = span({x2 1}). So Eig1 (T ) = Eig1 (T ) =
3 = dim(V ) and hance T is diagonalizable. Define = {x2 + x + 1, x2 1, x}.
Then
1 0 0
[T ] = 0 1 0 .
0 0 1
Of course isnt unique. Any permutation of the ordering will give you a diag-
onal matrix, perhaps with the diagonal entries permuted. BE CAREFUL: This
1
proof is done in a rather extemporaneous fashion. I would advise you to do it via
the standard approach of looking at the characteristic polynomial, eigenvalues,
etc.
Exercise 5.2.8: Suppose that A Mn (F) has two distinct eigenvalues, 1
and 2 , and that dim(E1 ) = n 1. Prove that A is diagonalizable.
Solution: Since dim(E2 ) 1, and we know that dim(E1 ) + dim(E2 ) n,
this forces dim(E2 ) = 1. Let Ei be the eigenspace of A corresponding to i .
Since E1 E2 = {0}, and under dimension considerations, we conclude that
V = E1 E2 . By Theorem 5.11, were done. (Note: Make sure you can prove
that E1 and E2 form a direct sum.)
Exercise 5.2.18a: Prove that if T and U are simultaneously diagonalizable
operators, then T and U commute.
Solution: Let be the basis that simultaneously diagonalizes T and U . Since
diagonal matrices commute with each other, we conclude that
[T U ] = [T ] [U ] = [U ] [T ] = [U T ] .
Since [T U ] = [U T ] , we conclude that T commutes with U . (Weve only proven
that with respect to this particular basis that their matrix representations com-
mute. Yet we are concluding something that is so much stronger. Why can we
do this? Make sure you know.)
Exercise 6.1.9 Let = {1 , ..., n } be a basis for a finite-dimensional in-
ner product space. (a) Prove that if hx, zi = 0 for all z , then x = 0. (b)
Prove that if hx, zi = hy, zi for all z , then x = y.
Solution: (a) Express x with respect to : x = a1 1 + an n . Then
Xn
hx|zi = h ai i |zi
i=1
n
X
= ai hi |zi (By property (i.) and (iii.) of the definition of h|i)
i=1
= 0 (By assumption).
(b) It suffices to prove that hxy|zi = 0, because we can simply use property (i.)
of the definition of the inner product to get what we want. But if hx y|zi = 0,
then by part (a) x y = 0 x = y. This completes our proof.
Exercise 6.1.12: Let {v1 , v2 , ..., vk } be an orthogonal set in V , and let a1 , a2 , ..., ak
be scalars. Prove that
Xk k
X
k ai vi k= |ai |2 kvi k2 .
i=1 i=1
2
Solution:
Xk Xk k
X
k ai vi k = h ai vi | aj vj i
i=1 i=1 j=1
k
X k
X
= ai hvi | aj vj i
i=1 j=1
k
X Xk
= ai h aj vj | vi i
i=1 j=1
k
X k
X
= ai aj hvj | vi i
i=1 j=1
k
X k
X
= ai aj hvi | vj i
i=1 j=1
k
X
= al al hvl | vl i
l=1
k
X
= |al |2 kvl k2
l=1
where used the orthogonality assumption.
Exercise 6.1.17: Let T be a linear operator on an inner product space V ,
and suppose that kT (x)k= kxk for all x. Prove that T is one-to-one.
Solution: Let x N (T ). Then ||x|| = ||T (x)|| = ||0||. By property (iv.) of
h | i we conclude x = 0. Therefore T is injective. This completes the proof.