### Linear Algebra: Example Sheet 2 of 4

```Michaelmas Term 2014
SJW
Linear Algebra: Example Sheet 2 of 4
1. (Another proof of the row rank column rank equality.) Let A be an m × n matrix of (column) rank r.
Show that r is the least integer for which A factorises as A = BC with B ∈ Matm,r (F) and C ∈ Matr,n (F).
Using the fact that (BC)T = C T B T , deduce that the (column) rank of AT equals r.
2. Write down the three types of elementary matrices and find their inverses. Show that an n × n matrix
A is invertible if and only if it can be written as a product of elementary matrices. Use this method to
find the inverse of


1 −1 0
0 0
1 .
0 3 −1
3. Let A and B be n × n matrices over a field F . Show that the 2n × 2n matrix
I
B
I B
C=
can be transformed into D =
−A 0
0 AB
by elementary row operations (which you should specify). By considering the determinants of C and D,
obtain another proof that det AB = det A det B.
4. (i) Let V be a non-trivial real vector space of finite dimension. Show that there are no endomorphisms
α, β of V with αβ − βα = idV .
(ii) Let V be the space of infinitely differentiable functions R → R. Find endomorphisms α, β of V which
do satisfy αβ − βα = idV .
5. Find the eigenvalues and give bases for the eigenspaces of the following complex matrices:






1 1 −1
1 1 −1
1 1 0
 −1 3 −1  .
 0 3 −2  ,
 0 3 −2  ,
−1 1 1
0 1 0
0 1 0
The second and third matrices commute; find a basis with respect to which they are both diagonal.
6. Let λ ∈ F. Consider the n × n matrix A with each diagonal entry equal to λ and all other entries 1.
How does the rank of A depend on λ? Evaluate det A.
7. Let V be a vector space, let π1 , π2 , . . . , πk be endomorphisms of V such that idV = π1 + · · · + πk and
πi πj = 0 for any i 6= j. Show that V = U1 ⊕ · · · ⊕ Uk , where Uj = Im(πj ).
Let α be an endomorphism on the vector space V , satisfying the equation α3 = α. Prove directly that
V = V0 ⊕ V1 ⊕ V−1 , where Vλ is the λ-eigenspace of α.
8. Let α be an endomorphism of a finite dimensional complex vector space. Show that if λ is an eigenvalue
for α then λ2 is an eigenvalue for α2 . Show further that every eigenvalue of α2 arises in this way. Are
the eigenspaces Ker(α − λι) and Ker(α2 − λ2 ι) necessarily the same?
9. (Another proof of the Diagonalisability Theorem.) Let V be a vector space of finite dimension. Show
that if α1 and α2 are endomorphisms of V , then the nullity n(α1 α2 ) satisfies n(α1 α2 ) ≤ n(α1 ) + n(α2 ).
Deduce that if α is an endomorphism of V such that p(α) = 0 for some polynomial p(t) which is a
product of distinct linear factors, then α is diagonalisable.
10. Let A be a square complex matrix of finite order - that is, Am = I for some m > 0. Show that A can be
diagonalised.
11. Let C be an n × n matrix over C, and write C = A + iB, where A and B are real n × n matrices. By
considering det(A + λB) as a function of λ, show that if C is invertible then there exists a real number
λ such that A + λB is invertible. Deduce that if two n × n real matrices P and Q are similar when
regarded as matrices over C, then they are similar as matrices over R.
[email protected]
-1 -
October 2014
12. Let A, B be n × n matrices, where n ≥ 2. Show that, if A and B are non-singular, then
(ii) det(adj A) = (det A)n−1 ,
(iii) adj (adj A) = (det A)n−2 A .
What happens if A is singular? [Hint: Consider A + λI for λ ∈ F.]

 n if r(A) = n
Show that the rank of the adjugate matrix is r(adj A) = 1 if r(A) = n − 1

0 if r(A) ≤ n − 2.
13. Let f (x) = a0 + a1 x + . . . + an xn , with ai ∈ C, and let C be the circulant matrix
 a
0
 an

 an−1
 .
 .
.
a1
a0
an
a2
a1
a0
a1
...
an 
. . . an−1 

. . . an−2  .
.. 
..

.
.
...
a0
a2 a3
Qn
Show that the determinant of C is det C = j=0 f (ζ j ), where ζ = exp(2πi/(n + 1)).
14. Let V denote the space of all infinitely differentiable functions R → R and let α be the differentiation
endomorphism f 7→ f 0 .
(i) Show that every real number λ is an eigenvalue of α. Show also that ker(α − λι) has dimension 1.
(ii) Show that α − λι is surjective for every real number λ.
15. Let α : V → V be an endomorphism of a real finite dimensional vector space V with tr(α) = 0.
(i) Show that, if α 6= 0, there is a vector v with v, α(v) linearly independent. Deduce that there is a
basis for V relative to which α is represented by a matrix A with all of its diagonal entries equal to 0.
(ii) Show that there are endomorphisms β, γ of V with α = βγ − γβ.
[email protected]
-2 -
October 2014
```