1 Matrices and Gaussian Elimination

Solving linear systems using elimination method

Recall: Common method to solve a linear system using Gaussian/Gauss-Jordan elimination:

1) Construct a matrix (called the augmented matrix) corresponding to the system

2) Using elementary row operations on matrices to reduce the augmented matrix to an upper triangu-
lar matrix.

3) Using back-substitution to solve the nal system.

Exercises:

1. Which of the following are linear equations in x1 , x2 and x3 ?

(a) x1 + 5x2 x3 = 1 (c) x1 2x2 + 13 x3 = 71/3

32
(b) x12 + x2 + 8x3 = 5 (d) x1 3x2 + 4 x3 = 2
5

2. The following matrices are the augmented matrices of systems of linear equatioins in variables
x1 , x2 , . . .. Write down the corresponding systems explicitly and solve them.

(a) (c)
1 6 0 0 4 2
1 0 0 5
0 1 0 0 0 1 0 3 1
2
0 0 0 1 5 2
0 0 1 4
0 0 0 0 0 0
(b) (d)

1 0 0 4 1 1 2 0 3 0 7
0 1 0 2 6 0 0 1 0 0 1
0 0 1 3 2 0 0 0 0 1 2

3. Solve the following linear systems using Gaussian or Gauss-Jordan (back-substitution) elimina-
tion. Write clearly all elementary row operations you used.

(a) (d)

x + y + 2z = 0 x 2y + z 4t = 1
2x + 4y 3z = 1 x + 3y + 7z + 2t = 2
3x + 6y 5z = 0. x 12y 11z 16t = 5.
(e)
(b)
2I1 I2 + 3I3 + 4I4 = 9
2x + 2y + 2z = 0
I1 2I3 + 7I4 = 11
2x + 5y + 2z = 1
3I1 3I2 + I3 + 5I4 = 8
8x + y + 4z = 1.
2I1 + I2 + 4I3 + 4I4 = 10.
(c) (f)
x + 2y t + w = 1 2u v =0
3y + z w = 2 u + 2v w =0
z + 7t = 1. v + 2w z = 0
w + 2z= 5.
1
 4. For which values of a will the following system have no solutions? Exactly one solution? In-
nitely many solutions?

x + 2y 3z = 4
3x y + 5z = 2
4x + y + ( a2 14)z = a + 2.

Operations on matrices
Recall:

1) Some special matrices: Identity matrices, zero matrices, symmetric matrices,...

2) Operations: addition, multiplication with a scalar, transpose of a matrix, multiplication.

3) General properties: Associativity, commutativity...

Exercises:
1. Suppose that A, B, C, D, E are matrices with the following sizes: A : 4 5; B : (4 5); C : 5 2;
D : 4 2; E : 5 5. Determine (if exists) the size of each of the following matrices:

(a) BA; (d) AB + B; (g) E T A;

(b) AC + D; (e) 2E( A + B);
(c) AE + B; (f) E( AC ); (h) ( A T + E) D.

2. Let

3 0   1 5 2 6 1 3
4 1 1 4 2
A = 1 2 , B = ,C = , D = 1 0 1 , E = 1 1 2 .
0 2 3 1 5
1 1 3 2 4 4 1 3

Compute the following (where possible)

(a) 2B C; (e) AC and CA;

(b) 3D 2E T ; (f) (C T B) A T and Tr ((C T B) A T );
(c) 3D T 2E) and Tr (3D 2E T ); (g) tr ( DD T ).
(d) AB and BA; (h) D T E T ( ED ) T .

3. Write down the 2 by 2 matrices A and B that have entries aij = i + j and bij = (1)i+ j . Multiply
them to nd AB and BA. Is the product of A and B commutative?

4. True or false? Give a specic counterexample when false.

(a) If columns 1 and 3 of B are the same, so are columns 1 and 3 of AB.
(b) If rows 1 and 3 of B are the same, so are rows 1 and 3 of AB.
(c) If rows 1 and 3 of A are the same, so are rows 1 and 3 of AB.
(d) ( AB)2 = A2 B2 .
5. Which of the following matrices are guaranteed to equal ( A + B)2

2
 (a) A2 + 2AB + B2 (c) ( A + B)( B + A)
(b) A( A + B) + B( A + B) (d) A2 + AB + BA + B2

6. By trial and error nd examples of 2 by 2 matrices such that

(a) A2 = I, A having only real entries.

(b) B2 = 0, although B = 0;
(c) CD = DC, not allowing the case CD = 0.
(d) EF = 0, although no entries of E or F are zero.
7. Suppose A commutes with every 2 by 2 matrix, and in particular
  
a b 1 0 0 1
A= commutes with B1 = , B2 = .
c d 0 0 0 0

Show that a = d and b = c = 0. Consequently, prove that if AB = BA for all 2 2-matrices B,

then A is a multiple of the identity.

8. In each part nd matrices A, X, B which express the given system of linear equations as a single
matrix equation AX = B. Solve these equations.

(a)

x1 3x2 + 5x3 = 7
9x1 x2 + x3 = 1
x1 + 5x2 + 4x3 = 0

(b)

x1 3x3 + x4 =7
5x1 + x2 8x4 =3
2x1 5x2 + 9x3 x4 =0
3x2 x3 + 7x4 = 2

9. Find the powers A2 , A3 , B2 , B3 , C2 , C3 . What are Ak , Bk and C k for a given k?

1  1
1
1 0 12
A= 2 2 ,B = , and C = AB = 21 .
1
2
1
2
0 1 2 12

3
Triangular factors and row exchanges
Recall: Reduce solving Ax to solving two triangular systems Lc = b and Ux = c, where A = LU and L
is a lower triangular matrix, and U is a upper triangular matrix.

1. Given the triangular factorization of A

1 1 1 1 0 0 1 1 1
A = 1 2 2 = 1 1 0 0 1 1 = LU
1 2 3 1 1 1 0 0 1

Solve the system Ax = b with a right-hand side b = (2, 2, 3).

2. Apply the elimination method to nd the LU decomposition for the following matrices

(a) (b) (c)

 3 1 1 1 1 1
2 1 1 3 1 1 4 4
8 7 1 1 3 1 4 8

3. Find the PA = LDU factorizations for

0 1 1 1 2 1
( a ) A = 1 0 1 ; ( b ) A = 2 4 2 .
2 3 4 1 1 1

Inverses of matrices
Recall:

1) Equivalent denitions: A is invertible if one the following equivalent conditions holds.

i) There exists B such that AB = BA = I;

ii) The equation Ax = 0 has unique solution 0.
2) Find the inverse of a matrix

Using elementary row transformations to bring [ A| I ] into [ I | A1 ]. (Gauss-Jordan.)

Exercises:

1. Show that if A and B are invertible matrices then

(a) A T is invertible and ( A T )1 = ( A1 ) T

(b) AB are invertible and ( AB)1 = B1 A1 .

2. Use the Gauss-Jordan method to invert the following matrices then solve the equations Ax = b
for b = (1, 2, 7).

(a) (b) (c)

1 0 0 2 1 0 1 1 1
A1 = 1 1 1 , A2 = 1 2 1 , A3 = 1 2 2 ,
0 0 1 0 1 2 1 2 3

4
 (d) (e) (f)
1 1 2
0 0 1 2 3 2 0 5 5 5
A4 = 0 1 1 , A5 = 4 2 2 0 , A6 = 1 1 1
5 5 10
1 1 1 0 0 1 1 4 1
5 5 10

3. True or false (with a counterexample if false and a reason if true):

(a) A 4 by 4 matrix with a row of zeros is not invertible.
(b) If A is invertible then A1 is invertible
(c) If A T is invertible, then A is invertible.
4. If a matrix A has row 1 + row 2 = row 3, show that A is not invertible:
(a) Explain why Ax = (1, 0, 0) cannot have a solution.
(b) Which right-hand sides (b1 , b2 , b3 ) might allow a solution to Ax = b?
(c) What happens to row 3 in elimination?
5. Find the inverse (in any legal way) of

(a) (b) (c)

0 0 0 1 a b 0 0 k 0 0 0
0 0 2 0 c d 0 0 1 k 0 0
A1 =
0
, A2 = A3 =
3 0 0 0 0 a b 0 1 k 0
4 0 0 0 0 0 c d 0 0 1 k

6. For which three numbers c is this matrix not invertible, and why not?

2 c c
A = c c c
8 7 c

7. Give examples of matrices A and B such that

(a) A + B is not invertible although A and B are invertible.
(b) A + B is invertible although A and B are not invertible.
(c) all of A, B, and A + B are invertible.
(d) In the last case use A1 ( A + B) B1 = B1 + A1 to show that C = B1 + A1 is also
invertible and nd a formula for C
8. Show that A2 = 0 is possible but A T A = 0 is not possible (unless A = zero matrix).
9. If the inverse of A2 is B, show that the inverse of A is AB. Thus, A is invertible whenever A2
invertible.
10. If A = A T and B = B T , which of the following matrices are certainly symmetric?
(a) A2 B2 ;
(b) ( A + B)( A B);
(c) ABA;
(d) ABAB.