Week-4

Mathematics for Data Science - 2

Basis of a vector space, Rank and dimension of a matrix
Practice Assignment

1. Match the vector spaces (with the usual scalar multiplication and vector addition as in
M3×3 (R) ) in column A with their bases in column B in Table : M2W4P1.

Vector space Basis

(Column A) (Column B)
   
 x y z  1 0 0 
a) V =  0 z x | x + y + z = 0, i) 0 1 0
y 0 0 0 0 1
  

)
and x, y, z ∈ R

   
 1 0 −1 0 1 −1 
b) V = {A | A ∈ M3×3 (R), ii) 0 −1 1  , 0 −1 0 
0 0 0 1 0 0
 

A is a diagonal matrix }

     
 1 0 0 0 0 0 0 0 0 
c) V = {A | A ∈ M3×3 (R), iii) 0 0 0 , 0 1 0 , 0 0 0
0 0 0 0 0 0 0 0 1
 

A is a scalar matrix }

Table : M2W4P1

Choose the correct option.

⃝ Option 1: a → i
 ⃝ Option 2: a → ii
⃝ Option 3: b → iii
⃝ Option 4: c → iii
⃝ Option 5: c → i
Solution:
a)
 
 x y z
)
V =  0 z x | x + y + z = 0, and x, y, z ∈ R
y 0 0

 
 x y −(x + y) )
=  0 −(x + y) x  | x, y ∈ R
y 0 0

    
 1 0 −1 0 1 −1 )
= x 0 −1 1  + y 0 −1 0  | x, y ∈ R
0 0 0 1 0 0

   
 1 0 −1 0 1 −1 
= span 0 −1 1  , 0 −1 0 
0 0 0 1 0 0
 
   
 1 0 −1 0 1 −1 
Thus 0 −1 1  , 0 −1 0  is a basis for V .
0 0 0 1 0 0
 

b)
V = {A | A ∈ M3×3 (R), A is a diagonal matrix}
 
 x 0 0
)
=  0 y 0 | x, y, z ∈ R
0 0 z

      
 1 0 0 0 0 0 0 0 0
= x 0 0 0 + y 0 1 0 + z 0 0 0 | x, y, z ∈ R}
0 0 0 0 0 0 0 0 1

     
 1 0 0 0 0 0 0 0 0 
= span 0 0 0 , 0 1 0 , 0 0 0
0 0 0 0 0 0 0 0 1
 
     
 1 0 0 0 0 0 0 0 0 
Thus 0 0 0 , 0 1 0 , 0 0 0 is a basis for V .
0 0 0 0 0 0 0 0 1
 

2
c)

V = {A | A ∈ M3×3 (R), A is a scalar matrix}

 
 x 0 0
)
= 0 x 0 | x ∈ R
0 0 x

  
 1 0 0
)
= x 0 1 0 | x ∈ R
0 0 1

 )
 1 0 0
= span 0 1 0
0 0 1


 )
 1 0 0
Thus 0 1 0 forms a basis for V .
0 0 1


3
2. Match the sets of vectors in column A with their properties of linear dependence or
independence in column B and the dimension of the vector spaces in column C spanned
by the sets.

Set of vectors Linear dependence Dimension of the vector space

or independence spanned by the set
(Column A) (Column B) (Column C)

a) {(1, 0, 0), (1, 1, 0), i) Linearly independent 1) 1

(1, 1, 1), (1, 0, 1)}

b) {(1, 0, −1), (−1, 2, 0), ii) Linearly dependent 2) 2

(−2, 0, 0)}

c) {(1, −1, 2), (−1, 1, −2), iii) Linearly dependent 3) 3

(2, −2, 4)}

d) {(1, 0, 1), (1, 1, 0), iv) Linearly dependent 4) 3

(0, −1, 1)}

Table : M2W4P2

Choose the correct option.

⃝ Option 1: a → ii → 4, b → i → 3
⃝ Option 2: a → ii → 3, b → i → 4
⃝ Option 3: c → iv → 1, d → iii → 2
⃝ Option 4: c → iv → 2, d → iii → 1
Solution:
a) Let S1 = {(1, 0, 0), (1, 1, 0), (1, 1, 1), (1, 0, 1)}. Four vectors cannot be linearly indepen-
dent in R3 . Hence S1 is linearly dependent. Observe that (1, 0, 1) − (1, 0, 0) + (1, 1, 0) =
(1, 1, 1).
Now consider the subset S1∗ = {(1, 0, 0), (1, 1, 0), (1, 0, 1)} of S1 . Clearly, span(S1∗ ) =
1 0 1
span(S1 ). S1 is also linearly independent because 1 0 0 = 1 ̸= 0. Thus dim(span(S1∗ )
∗

1 1 0
= dim(span(S1 ) = 3.

4
 Hence a → (ii) → 4.

1 0 −1
b) Let S2 = {(1, 0, −1), (−1, 2, 0), (−2, 0, 0)}. S2 is linearly independent since −1 2 0 =
2 0 0
−4 ̸= 0. Thus dim(span(S2 ) = 3.
Hence b → (i) → 3.

c) Let S3 = {(1, −1, 2), (−1, 1, −2), (2, −2, 4)}. Since −1(1, −1, 2) = (−1, 1, −2) and
2(1, −1, 2) = (2, −2, 4), span(S3 ) = span{(1, −1, 2)}. Hence dim(span(S3 )) = dim(span{(1, −1, 2)})
= 1.
Hence c → (iv) → 1.

d) Let S4 = {(1, 0, 1), (1, 1, 0), (0, −1, 1)}. Since, (1, 0, 1) − (1, 1, 0) = (0, −1, 0) and
since (1, 0, 1) ̸= α(1, 1, 0) for any α ∈ R, {(1, 0, 1), (1, 1, 0)} is linearly independent. Also
span(S4 ) = span{(1, 0, 1), (1, 1, 0)}. Hence dim(span(S4 )) = dim(span{(1, 0, 1), (1, 1, 0)})
= 2.
Hence d → (iii) → 2.

3. Which of the following options are true?

⃝ Option 1: The number of linearly independent vectors in a vector space can
be more than the dimension of the vector space.
 
1 −1
⃝ Option 2: Dimension of the vector space spanned by the vectors ,
    1 1
−1 1 −1 1
, and is 2.
−1 1 1 1
  
1 −1 −1 1
⃝ Option 3: Dimension of the vector space spanned by the vectors , ,
  1 1 −1 −1
−2 2
and is 1.
−2 −2
⃝ Option 4: The number of spanning vectors can be more than the dimension
of the vector space.

5
 Solution:
Option 1: From the definition of dimension of vector space, the number of linearly
independent vectors in a vector space V can be at most dim(V ). Hence option 1 is
incorrect.
     
1 −1 −1 1 −1 1
Option 2: Let S = { , , }.
1 1 −1 −1 1 1
       
1 −1 −1 1 −1 1 0 0
Suppose α +β +γ = .
1 1 −1 −1 1 1 0 0
   
α − β − γ −α + β + γ 0 0
Then we have = .
α−β+γ α+β+γ 0 0

α−β−γ =0 (1)
−α + β + γ =0 (2)
α−β+γ =0 (3)
α+β+γ =0 (4)

Solving these equations, we get α = β = γ = 0. Hence, S is a linearly independent set

and dim(span(S)) = 3. Option 2 is incorrect.
     
1 −1 −1 1 −2 2
option 3: Let S = { , , }.
1 1 −1 −1 −2 −2
       
1 −1 −1 1 1 −1 −2 2
Observe that −1 = and −2 = . Thus span(S) =
  1 1 −1 −1 1 1 −2 −2
1 −1
span{ }. Thus dim(span(S)) = 1 and option 3 is correct.
1 1
Option 4: If S is a spanning set for the vector space V , then S must contain at least
dim(V ) vectors. Note that any superset of a spanning set is a spanning set and hence the
number of spanning vectors can be more than the dimension of V . Option 4 is correct.

4. Let A be a nonzero 3 × 3 matrix. Which of the following options are true?

⃝ Option 1: The rank of A must be at least 1.
⃝ Option 2: The rank of A may be 0.
⃝ Option 3: The rank of A must be strictly less than 3.
⃝ Option 4: If the rank of the matrix is 2, then the dimension of the vector
space spanned by the vectors corresponding to each column of A, must be 2.

6
 ⃝ Option 5: If the rank of the matrix is 2, then there exists one vector corre-
sponding to a column of A, which can be expressed as a linear combination of
the vectors corresponding to each of the remaining columns of A.
Solution: Given, A is a non-zero matrix. So, at least one of the rows of A must be
non-zero and hence rk(A) ≥ 1. Hence, option 1 is correct and option 2 is incorrect.
Now, A is a 3 × 3 matrix. A may be invertible or non-invertible. If A is invertible, then
rk(A) = 3. Hence option 3 is incorrect.
For any matrix A, rowrank(A) = columnrank(A) = rk(A). Thus if the rank is 2, then
dimension of row space and column space are equal to 2. Hence, option 4 is correct.
If the rank of the matrix is 2, then the number of linearly independent columns of A is 2.
Hence, one column can be expressed as a linear combination of the other two columns.
Option 5 is correct.

 
1−m 2 4
5. Suppose A =  3 0 0 . For what value of m is the rank of A at most 2?
m m −2
[Answer: −1]
Solution:
 
1−m 2 4
Given, A =  3 0 0
m m −2
We know that the nullity of A is at least 1 if and only if the matrix A is singular i.e.,
det(A) = 0.
1−m 2 4
3 0 0 = 0 if and only if (1 − m)(0 × −2 − 0 × m) − 2(3 × −2 − 0 × m) + 4(3 ×
m m −2
m − 0 × m) = 0, this gives m = −1.
For m = −1 the matrix A is singular which implies that the nullity of A is at least 1
and hence rank is atmost 2.

 
0 −1 a
6. If rank of the matrix 2 0 −4 is 2 then find the value of a.
3 −9 −6

7
 Solution:
 
0 −1 a
Let A = 2 0 −4.
3 −9 −6
(2, 0, −4) ̸= α(3, −9, −6) for any α ∈ R. Thus the set {(2, 0, 4), (3, −9, −6)} is linearly
independent. This implies rk(A) ≥ 2. Given that rk(A) = 2. This is possible if and only
if det(A) = 0.
0 −1 a
det(A) = 2 0 −4 = 0 if and only if −(−1)(2 × −6 − 3 × −4) + a(2 × −9 − 3 × 0) = 0.
3 −9 −6
That is (−12 + 12) + a(−18) = 0 and hence a = 0.

 
a 1
7. Find out the value of a for which the matrix will be in the spanning set of
    0 −1
1 0 0 1
the matrices and in M2×2 (R) with usual matrix addition and scalar
0 −1 0 0
multiplication.
     
a 1 1 0 0 1
Solution: For to be in the spanning set of and , we must
  0 −1   0 −1  0 0  
a 1 1 0 0 1 a 1
have to be a linear combination of and . That is =
 0 −1     0 −1 0 0 0 −1
1 0 0 1 α β
α +β =
0 −1 0 0 0 −α

=⇒ α = a, β = 1, α = 1 =⇒ a = 1

8
 Comprehension Type Question:

Suppose in a village there are four farmers A, B, C and D, each owning 1 acre of land. They
cultivate paddy, pulses and/or sugarcane in their lands as follows: Farmer A uses 50% of his
land for paddy, 30% for pulses and the remaining for sugarcane. Farmer B uses 40% of her
land for paddy and she divides her remaining land equally for pulses and sugarcane. Farmer
C uses the whole land for paddy only, and Farmer D uses 30% for paddy, 30% for pulses and
the remaining for sugarcane. Using the above information, answer the following questions.

8. Suppose the area used by a farmer for different crops is denoted by a row vector. Let S
be the span of the resulting four row vectors . Choose the correct set of options.
⃝ Option 1: The row vectors corresponding to the area used for different crops
by Farmer D can be obtained as a linear combination of the row vectors cor-
responding to the area used for different crops by Farmers A, B, and C.
⃝ Option 2: The row vectors corresponding to the area used for different crops
by Farmer D can be obtained as a linear combination of the row vectors cor-
responding to the area used for different crops by Farmer A and Farmer B.
⃝ Option 3: The row vectors corresponding to the area used for different crops
by Farmer C can be obtained as a linear combination of the row vectors cor-
responding to the area used for different crops by Farmer A and Farmer B.
⃝ Option 4: The row vectors corresponding to the area used for different crops
by Farmer D can be obtained as a linear combination of the row vectors cor-
responding to the area used for different crops by Farmer A and Farmer C.
Solution:

paddy pulses sugarcane

5 3 2
A 10 10 10
4 3 3
B 10 10 10
10 0 0
C 10 10 10
3 3 4
D 10 10 10

Option 1:
3 3 4 5 3 2 4 3 3
( 10 , 10 , 10 ) = −( 10 , 10 , 10 ) + 2( 10 , 10 , 10 ) = −1A + 2B + 0C
The row vector corresponding to farmer D can be obtained as a linear combination of
those of A, B, C.
Option 2: From the above representation, we get that the row vector corresponding to
farmer D can be obtained as a linear combination of those of A and B.

9
 Option 3: If the row vector corresponding
 5 4 10  to farmers A, B, and C can be expressed as
10 10 10
3 3 0 
columns of the matrix, then  10 10 10
. The determinant is non-zero so the vectors
2 3 0
10 10 10
are linearly independent and hence the row vector of farmer C cannot be written as a
linear combination of those of A and B.
Option 4: Similar to option 3 by choosing A, C, D.

9. Let S be the vector space defined in the previous question, with the usual addition and
scalar multiplication on R3 . Which of the following sets will not be a basis of S?
⃝ Option 1: {(5, 3, 2)}
⃝ Option 2: {(5, 3, 2), (4, 3, 3)}
⃝ Option 3: {(5, 3, 2), (4, 3, 3), (10, 0, 0)}
⃝ Option 4: {(5, 3, 2), (4, 3, 3), (3, 3, 4)}
⃝ Option 5: {(5, 3, 2), (4, 3, 3), (10, 0, 0), (3, 3, 4)}

Solution:
Since the row vectors corresponding to farmers A, B, C are linearly independent (option
3 of question 8) {(5, 3, 2), (4, 3, 3), (10, 0, 0)} is a linearly independent set and since D is a
linear combination of these vectors {(5, 3, 2), (4, 3, 3), (10, 0, 0)} is a basis of S. Hence the
dimension of S is 3.
Options 1,2,& 5 cannot be a basis of S (since dimension is not 3).
The set in option 4 was already shown to be linearly dependent. Hence it cannot be a
basis.

10