NATIONAL UNIVERSITY OF SINGAPORE
Department of Mathematics
2025-2026 (Semester 1) MA1513 Lecture Problem Set 3 (Solutions)
1. (i) “Standard Approach” Set up the vector equation c1 v 1 + c2 v 2 + · · · + ck v k = 0
for each of the given sets of vectors and convert them into linear systems. Then
use Gaussian elimination to solve for c1 , c2 , · · · , ck .
(a) c1 (1, 0, −1) + c2 (−1, 2, 3) = (0, 0, 0)
1 −1 0 1 −1 0
G.E.
0 2 0 −→ 0 2 0 .
−1 3 0 0 0 0
As there is no non-pivot column, the system has only the trivial solution, and
the vectors are linearly independent.
(b) c1 (1, 0, −1) + c2 (−1, 2, 3) + c3 (0, 3, 0) = (0, 0, 0)
1 −1 0 0 1 −1 0 0
G.E.
0 2 3 0 −→ 0 2 3 0 .
−1 3 0 0 0 0 −3 0
As there is no non-pivot column, the system has only the trivial solution, and
the vectors are linearly independent.
(c) c1 (1, 0, −1) + c2 (−1, 2, 3) + c3 (0, 3, 3) = (0, 0, 0)
1 −1 0 0 1 −1 0 0
G.E.
0 2 3 0 −→ 0 2 3 0 .
−1 3 3 0 0 0 0 0
As there is a non-pivot column, the system has non-trivial solutions, and the
vectors are linearly dependent.
(d) c1 (1, 0, −1) + c2 (−1, 2, 3) + c3 (0, 3, 0) + c4 (1, −1, 1) = (0, 0, 0)
1 −1 0 1 0 1 −1 0 1 0
G.E.
0 2 3 −1 0 −→ 0 2 3 −1 0 .
−1 3 0 1 0 0 0 −3 3 0
As there is a non-pivot column, the system has non-trivial solutions, and the
vectors are linearly dependent.
“Shortcut” methods
(a) The two vectors (1, 0, −1), (−1, 2, 3) are not scalar multiple of each other. So
they are linearly independent.
1 −1 0
(b) We can form the 3 × 3 determinant 0 2 3 which is −6 ̸= 0. So the
−1 3 0
three column vectors are linearly independent.
1 −1 0
(c) We can form the 3 × 3 determinant 0 2 3 which is 0. So the three
−1 3 3
column vectors are linearly dependent.
1
(d) There are four vectors, and the vectors are from R3 . So these vectors are
linearly dependent.
(ii) Since the dimension of R3 is 3, any basis for R3 must contain three vectors,
and in fact, any three vectors that are linearly independent will form a basis fo R3 .
As (a) and (d) have 2 an 4 vectors respectively, they do not form bases for R3 .
(c) has 3 vectors but they are linearly dependent. So it cannot be a basis for R3
as well.
(b) is a set of 3 linearly independent vectors. So it is indeed a basis for R3 .
2. (a) The augmented
matrix of the
homogeneous system is already in row echelon
1 3 −1 2 0
form: .
0 −3 1 0 0
By back substitution, we get the general solution
x1 −2t 0 −2
x2 s 1 0
= 3 = s 3 + t where s, t ∈ R.
x3 s 1 0
x4 t 0 1
So {(0, 13 , 1, 0), (−2, 0, 0, 1)} is a basis for the solution space, and the dimen-
sion of the solution space is 2.
(b)
After solving the homogeneous
system using Gaussian
elimination:
1 3 −1 2 0 1 3 −1 2 0
G.E.
0 −3 1 0 0 −→ 0 −3 1 0 0
1 0 0 −1 0 0 0 0 −3 0
we get a general solution
x1 0 0
x2 t 1
= 3 = t 3 where t ∈ R.
x3 t 1
x4 0 0
So {(0, 13 , 1, 0)} is a basis for the solution space, and the dimension of the
solution space is 1.
3. (a) By some sketching on the xy-plane and simple trigonometry, we will√have √
the
2 2
vector (1, 0, 0) (the unit vector along the x-axis) being rotated to ( 2 , 2 , 0)
and√ the
√
vector (0, 1, 0) (the unit vector along the y-axis) being rotated to
2 2
(− 2 , 2 , 0), while the vector (0, 0, 1) (the unit vector along the z-axis) re-
mains unchanged. So these three vectors form the basis S for the new coor-
dinate axes.
(b) We need to find a, b, c such that
√ √ √ √
2 2 2 2
a( , , 0) + b(− , , 0) + c(0, 0, 1) = (1, 1, 1).
2 2 2 2
√
After solving, we get a = 2, b = 0, c = 1. These give the coordinates of v
relative to the new axes.
2
(c) By writing the vectors in column forms, the vector equation in part (b) with
the solution substituted gives:
√2 √2
√ √2
−√ 2 0 1
2 2 +0
2 2 + 0 = 1
2
0 0 1 1
which can be further expressed in matrix form:
√2 √
2
√
2
√
−√ 2
0 2 1
2 2
0 0 = 1
2 2
0 0 1 1 1
Moving the 3 × 3 matrix to the right hand side, we get (v)S = M v where
√2 √
2
−1 √2 √2
√2
−√2
0 2√ √2
0
M= 2 2 = −2 2 2
2 2
0
2
0.
0 0 1 0 0 1
1 −1 0 1 1 −1 0 1
0 1 −1 1 0 1 −1 1
G.E.
4. (a) (A | b) = −1 0
−→
1 1 0 0 3 −2
1 1 1 1 0 0 0 3
which gives an inconsistent system.
(b) We solve the system AT Ax = AT b, which is given by
3 0 0 x 1
0 3 0 y = 1
0 0 3 z 1
The solution of this new system x = 31 , y = 13 , z = 13 gives the least squares
solution for Ax = b.
(c) Note that the linear span is the column space of A. So the required projection
p is given by Ax0 where x0 is the least squares solution in part (b) written
as a
column matrix.
1 −1 0 1 0
0 1 −1 31 0
p= −1 0
=
1 31 0
1 1 1 3 1
(d) Since p is the projection of b on the subspace:
span{(1, 0, −1, 1), (−1, 1, 0, 1), (0, −1, 1, 1)},
1
1
we have b − p = 1 orthogonal to this subspace.
0
In particular, this vector is orthogonal to (1, 0, −1, 1), (−1, 1, 0, 1), (0, −1, 1, 1).
3
5. (i) The row echelon forms are obtained from the original matrices by performing
Gaussian elimination. Without additional information, we will not be able to
find the original matrices A and B.
(ii) Since we are able to find bases for the row spaces of A and B by using the
non-zero rows of their respective row echelon forms RA and RB , we can find
these row spaces by taking the linear spans of the respective basis:
row space of A = span{(1, 0, 1, 2), (0, 0, 1, 1)};
row space of B = span{(0, 1, 0, 1), (0, 0, 1, −1), (0, 0, 0, 1)}.
(iii) There are two pivot columns in RA . So a basis for the column space of A
has two vectors, given by the first and third columns a1 and a3 of A. Since
we do not know A, we cannot find the column space of A.
There are three pivot columns in RB . So a basis for the column space of
B has three vectors, given by the second, third and fourth columns b2 , b3 , b4
of B. Though we do not know B, we know any three linearly independent
vectors will span the entire 3-space R3 . In other words,
the column space of B = span{b2 , b3 , b4 } = R3 .
A basis for row space of A: (1, 0, 1, 2), (0, 0, 1, 1), (0, 0, 0, 0)
A basis for row space of B: (0, 1, 0, 1), (0, 0, 1, −1), (0, 0, 0, 1)
A basis for column space of A: unknown
A basis for column space of B: though the columns of B are unknown, this
3
column
space is the same as R itself. So we can use the standard basis
1 0 0
0 , 1 , 0.
0 0 1