You are on page 1of 11

1

Linear Algebra
IMPORTANT CONCEPTS, FORMULAE AND THEOREMS

1. Singular Square Matrix: |A| = 0.


2. Non-singular Square Matrix |A| 0.
3. Elementary Row Operations for Matrices
(a) Interchanging two rows
(b) Multiplying each element of a row by the same non-zero scalar quantity
(c) Adding or subtracting corresponding elements from those of another row. These operations
are called elementary row operations. There is a corresponding set of three elementary
column operations that can be used to form column equivalent matrices.
4. Consistency of a Set of n Equations in n Unknowns with Coefficient Matrix A and
Augmented Matrix Ab.
(a) Consistent if rank of A = rank of Ab.
(b) Inconsistent if rank of A < rank of Ab.
5. Existence of Solution of a Non Homogeneous System of Linear Equation
Unique solution
No solution
Many solution

r ( A) = r ( Ab ) = n
r ( A) r ( Ab )
r ( A) = r ( Ab ) < n

Uniqueness of solutions n equations with m unknowns


(a) Rank of A = rank of Ab = n unique solutions
(b) Rank of A = rank of Ab = m < n infinite number of solutions
(c) Rank of A < rank of Ab no solution

GATE Engineering Mathematics

6. Solution of Sets of Equations


(a) Inverse matrix method Ax = b; x =A1 b
To find A1
(i) Evaluate |A|
(ii) Form C, the matrix of cofactors of A
(iii) Write CT, the transpose of C
(iv) A1 =

1 T
adj( A)
C , i.e A 1 =
det( A)
A

(b) Row transformation method Ax = b; Ax = Ib


(i) Form the combined coefficient matrix x A = I z
(ii) Row transformations to convert to [ I  A ~ ] A , will be inverse of A.
(iii) Then solve x = A1 b.
(c) Gaussian elimination method Ax = b
(i) Form augmented matrix [Alb]
(ii) Operate on rows to convert to [U l b] where U is the upper- triangular matrix.
(iii) Expand from bottom row to obtain x.
(d) Triangular decomposition method Ax = b
Write A as the product of upper and lower triangular matrices.
A = LU, L(Ux) = b. Put U xy .. Ly = b
(i) Solve Ly = b to obtain y.
(ii) Solve Ux = y to obtain x.
7. Matrix Polynomial
Let f ( x ) = a 0 x n + a1 x n 1 + a2 x n 2 + ....... + an 1 x n 1 + an be a polynomial and let A be a square
matrix of order n. Then f ( A) = a 0 An + a1 An 1 + a2 An 2 + ... + an 1 An 1 + an I n is called
matrix polynomial of A.
8. Characteristics Equation of a Matrix
Let A be A = (aij)nn be a square matrix. Then the matrix |A | is called characteristic
polynomial of matrix A and |A | = 0 and is called characteristic equation of A. The roots of
the characteristic equation are called eigenvalues of A.

Linear Algebra

1 2
Example: Find the eigenvalues of A =

0 3

The characteristic polynomial is A I =


The characteristic equation is A I =

1
2
= 2 4 + 3
0
3

1
2
=0
0
3

Therefore, A I = (1 )(3 ) = 0
The eigenvalues are 1 = 1, 2 = 3.
9. Eigenvalues and Eigenvectors Ax = x
Sets of equations of form Ax = x , where A = coefficient matrix,
x = Column matrix, = scalar quantity.
Equations become = ( A I) x = 0.
For non-trivial solutions, | A I| = 0 is the characteristic equation and gives the eigenvalues
of A. Substitution of each eigenvalue gives a corresponding eigenvector.
10. The CayleyHamilton Theorem
State: Every square matrix satisfies its own characteristic equation.
Let Pn ()be the characteristic polynomial of an arbitrary n n square matrix A. Then A
satisfies its own characteristic equation, and so is a solution of the matrix polynomial equation
Pn (A) = 0.
Let f ( x ) = a 0 x n + a1 x n 1 + a2 x n 2 + ....... + an 1 x n 1 + an be a polynomial and let A be a square
matrix of order n. Then f ( A) = a 0 An + a1 An 1 + a2 An 2 + ... + an 1 An 1 + an I n is called
matrix polynomial.
11. Nullity of a Matrix
If A is square matrix of order n and (A) is rank of the matrix A. Then nullity of the matrix is
given by n (A).The nullity is denoted by N(A). i.e. N(A) = n (A).
Hence rank + nullity = n
12. Matrix Transformations
(a) U = TX, where T is a transformation matrix, transforms a vector in the x-y plane to a
corresponding vector in the u-v plane. Similarly, X = T1U converts a vector in the u-v
plane to a corresponding vector in the xy plane.
(b) Rotation of axes

GATE Engineering Mathematics

13. Idempotent and Nilpotent Matrices


A is idempotent if A2 = A , and B is nilpotent if Bm = O for some positive integer m.
14. Definition of Different Matrix and Nature of Eigenvalues
Matrix
Symmetric
Skew
Symmetric

Definition

Hermitian

a ij = a ji

Skew
Hermitian

a ij = a ji

aij = aji
aij = aji

Diagonal elements are purely real


Diagonal elements are purely imaginary
or zero
If A, B are orthogonal matrices, then

Orthogonal

AA = I = A A
T

Unitary

Properties
Diagonal elements are non zero
Diagonal elements are zero

AA = I = AT A

A 1 , AT , AB are orthogonal, A1 = AT

If A,B are unitary matrices, Then

A 1 , AT , AB are orthogonal, A 1 = AT

Eigenvalues
Purely real
Purely
imaginary or
zero
Purely real
Purely
imaginary
Real or
complex
conjugate in
pair and have
absolute value
1.
Unit modulus

15. Properties of Eigenvalues


Let A be an n n matrix with eigenvalues i, i = 1, 2, ..., n. Then
(i) Trace of A = 1 + 2 + ..... + n.
(ii) Determinant of A = 1 2 .... n.
(iii) Eigenvalues of Ak are 1k , 2k , 3k ,......., kn .
(iv) If A is Hermitian, every eigenvalue of A is purely real.
(v) If A is skew Hermitian, every eigenvalue of A is purely imaginary or zero.
(vi) Every eigenvalue of a Unitary matrix has absolute value | | = 1.
16. Rank of a Matrix
The highest order non zero minor is called the rank of a matrix i.e. the rank of a matrix A is the
maximum number of linearly independent row vectors of A. It is denoted by rank A or R (A).
The number r is called the rank of an m n matrix A if
(i) Every square sub matrix of order r + 1 or more is singular.
(ii) There exist at least one sub matrix of r is non singular.
Thus, the rank of a matrix is the highest order non zero minor.
17. Properties of a Rank of a Matrix
(i) Only a null matrix has rank zero.
(ii) rank (Amn) min (m, n).

Linear Algebra

(iii) If Det (A) = 0, then rank (A) 1


(iv) rank(Amn) = n iff its columns are linearly independent.
(v) rank(A) = rank(AT)
(vi) rank(A) = maximum number of linearly independent columns (or rows) of A.
(vii) rank(A) is the dimension of the rank of A.
(viii) rank (Ann) + nullity (Ann) = n
(ix) det (Ann) = 0 iff rank(Ann) < n.
(x) rank(A + B) <= rank(A) + rank(B)
(xi) rank(AAH) = rank(AHA) = rank(A)
(xii) [X: non-singular]: rank(XA) = rank(AX) = rank(A)
(xiii) rank (DIAG(A, B, ..., Z)) = sum(rank(A), rank(B), ..., rank(Z))
(xiv) If A * denotes the conjugate transpose of A (i.e., the adjoint of A), then
Rank (A) = Rank (AT) = Rank (A*) = Rank (AAT) = Rank (AA*)
18. Relation between Eigenvalues of A, A2 , A3, .....
If 1, 2, 3 be the eigenvalues of a matrix A, then
Matrix

Eigenvalues

1, 2 , 3

A2

21 , 22 , 23

An

n1 , n 2 , n 3

A1

11 , 12 , 13

AT

1 , 2 , 3

19. Some Important Results


(i) Every skew symmetric matrix of odd order is singular.
(ii) If A is skew-symmetric matrix then A2 is symmetric.
(iii) A and its transpose AT have the same rank.
(iv) A homogeneous linear system with fewer equations than unknowns has always nontrivial solutions.
(v) The two eigenvalues of 2 2 matrices is =

trace [( trace ) 2 4 det]


2

(vi) The inverse matrix gives AA1 = I and A1 A = I


(vii) The Gauss = Jordon method solves AA1 = I to find the n column of A1 . The augmented
matrix [A I] is row reduced to [I A1]

GATE Engineering Mathematics

(viii) The inverse of AB is the reverse product of B1 A1 and (ABC)1 = C1 B1 A1


(ix) For a given matrix A the solution set of the homogeneous system Ax = 0 is a vector
space, called the null space of A and its dimension is called the nullity of A.
The basic relation rank A + nullity A = Number of columns of A.
Properties of determinants
A determinant det A has the following properties:
(i) If any row or column of a determinant det A only contains zero elements, then determinants
det A = 0.
(ii) If A is a square matrix with the transpose A, then det A = det A.
(iii) If each element of a row or column of a square matrix A is multiplied by a constant k,
then the value of the determinant is k det A.
(iv) If two rows (or columns) of a square matrix are interchanged, the sign of the determinant
is changed.
(v) If any two rows or columns of a square matrix A are proportional, then det A = 0.
(vi) The addition of a multiple of a row (or column) of a determinant to another row (or
column) of the determinant leaves the value of the determinant unchanged.
(vii) Let A and B be two n n matrix, then det(AB) = det A det B.

EXERCISES
1. Obtain the eigenvalues of the matrix:

A=

(A) 1, 2, 2, 1

49
0 2 43 94
0 0 2 104

0 0 0 1

1 2 34

(B) 1, 2, 1, 2

(C) 0

(D) 3, 4, 3, 4

1 1 3
2. The distinct eigenvalues of the matrix M = 1 5 1 are

3 1 1
(A) 0 and 1

(B) 1 and 1

(MA-11)

(C) 1 and 2

(D) 0 and 2

k 1 2
3. If the nullity of the matrix 1 1 2 is 1, then the value of k is

1 1 4
(A) 1

(B) 0

(C) 1

(D) 2

Linear Algebra

4. If 33 a real skew symmetric matrix has an eigenvalue 2i, then one of the remaining eigenvalues
is

1
1
(B)
(C) 1
(D) 0
(CS -10)
2i
2i
5. The eigenvalues of a 2 2 matrix are 2 and 3. The eigenvalues of the matrix (X + I)1
(X + 5I) is
(A)

(A) 3, 1

(B) 2, 4

(C) 1, 2

(D) 1, 3

6 2 2
6. If the product of two eigenvalues of a matrix 2 3 1 is 14, then the third eigenvalue is

3 1 3
(A) 1

(B) 2

(C) 3

(D) 4

7. Which of the following matrix is not diagonalizable?


1 1
(a)
1 2

1 0
(b)
3 2

0 1
(c)
1 0

1 1
(d)
0 1

(A) a, b, c

(B) a, b

(C) b, c

(D) none

1 2 3
2 0 0
8. If A = 0 2 1 , B = 0 1 3 , then determinant of AB is equals to

0 0 2
0 0 2
(A) 4

(B) 8

(C) 16

(D) 32

3 4
9. If A =
, then An is
1 1

( 4 )n

( 1)n

1 + 2n 4n
(A)
1 2n
n

3n
(B)
n

1 + 2n 4n
(C)
1 + n 1 n

1 + 2n 4n
(D)
1 + n 1 2n

3 4 0
10. Let N = 4 3 0 , then N is

0 0 1
(A) Non invertible

(B) Symmetric

(C) Skew symmetric

(D) Orthogonal

GATE Engineering Mathematics

11. A matrix M has eigenvalue 1 and 4 with corresponding eigenvectors (1, 1)T and (2,1)T
respectively. Then M is
4 8
(A)
5 9

9 8
(B)
5 4

2 2
(C)
1 3

3 2
(D)
1 2

12. If A and B be real symmetric matrices of size n n, then.....


(A) A = A1

(B) AB = BA

(C) ( AB )T = BA

(D) AAT = I

13. If A and B are same matrices of size n n then which of the following statement is not true?
(A) det(kA) = kndet(A)

(B) det(AB) = det(A)det(B)

(C) det(A+B) = det(A) + det(B)

(D) det(AT) = 1/det(A1)

14. If AT = A1, where A is a real matrix, then A is


(A) Orthogonal

(B) Symmetric

(C) Normal

(D) Hermitian

15. The system of equations, 2x + 4y = 10, 5x + 10y = 25


(A) Has only two solutions

(B) Has infinite solutions

(C) Has only one solution

(D) Has no unique solution

0 0
16. The eigenvector(s) of the matrix 0 0 0

0 0 0
(A) (0, 0, )

(B) (, 0, 0)

, 0 is (are):

(C) (0, 0, 1)

(CS-93)
(D) (0, , 0)

17. Let A and B be real symmetric matrices of size n n. Then which one of the following is true?
(B) A = A1

(A) AA' = 1

(C) AB = BA

(D) (AB)' = BA (CS-94)

0 0 3
18. The rank of matrix 9 3 5 is

3 1 1
(A) 0

19. If A = i

(A) 0

(B) 1
0
1 + i 3
2
1 + 2i

(CS-94)

(C) 2

(D) 3

, then the trace A100 is


0

1 i 3

2
0

(B) 1

(C) 2

(D) 3

Linear Algebra

a 2 ab ac

20. Consider the matrix M = ab b 2 bc where a, b and c are nonzero real numbers, then

2
ac bc c
the matrix has
(A) Three nonzero real eigenvalues

(B) Complex eigenvalues

(C) Two nonzero eigenvalues

(D) Only one non zero eigenvalue

21. The rank of the following (n + 1) (n + 1) matrix, where a is a real number is:

(CS-95)

1 a a2  an

1 a a2  an

  

  


1 a a 2  a n

(A) 1

(B) 2

(C) n

(D) Depends on the value of a

22. Let A be a 33 matrix. Suppose that the eigenvalues of A are 1, 1, 0 with respective eigenvectors
(1, 1, 0)T, (1, 1,2)T, and (1, 1, 1)T. Then 6A equal to

1 5 2

(A) 5 1 2
2 2 2

1 0 0

(B) 0 1 0
0 0 0

1 5 3

(C) 5 1 3
3 3 3

3 9 0

(D) 9 3 0
0 0 6

23. Let A be n n complex matrix having characteristic polynomial f ( x) = x n + an 1 x n 1 +


an 2 x n 2 + .... + a1 x + a0 , then det A is

(A) an 1

(B) a0

(C)

( 1)n an1

(D)

24. The system AX 0 , where A is an n n matrix find true statement


(A) May not have a nonzero solution
(B) Always has a nonzero solution
(C) Always has at least two linearly independent solutions
(D) Always has at least n linearly independent solutions

( 1)n a0

10

GATE Engineering Mathematics

1
1
25. The rank of matrix
2

1
(A) 3

0
2
is:
2 0

0 1

1
1

(B) 4

(C) 2

2
2
26. Number of linearly independent eigenvectors of the matrix
0
0

(A) 1

(B) 2

1
0
27. The rank of matrix M =
1
0

(A) 1

(D) 1
2 0 0
1 0 0
is:
0 3 0
0 1 4

(C) 3

(D) 4

(C) 3

(D) 4

1 0 0
0 1 1
is:
0 1 0
1 0 1

(B) 2

28. Let A be n n matrix which is both hermitian and unitary. Then


(A) A2 = I
(B) A is real
(C) The eigenvalues of A are 0, 1,1
(D) Characteristic and minimal polynomials of A are the same.
29. Let M be a 3 3 complex hermitian matrix which is unitary. Then the distinct eigenvalues of
M are
(A) i

(B) 1 i

(C) 1

(D)

1
(1 i )
2

30. Let M be a skew symmetric, orthogonal real matrix, then only possible eigenvalues are
(A) 1

(B) i

(C) 0

(D) 1, i

31. If M is a 7 5 matrix of rank 3 and N is a 5 7 matrix of rank 5, then rank (MN) is


(A) 1

(B) 5

(C) 3

(D) 2

32. If P is a m n matrix of rank n and Q is a n p matrix of rank p, then rank (PQ) is


(A) n

(B) p

(C) np

(D) n + p

Linear Algebra

11

33. Let P and Q be square matrices such that PQ = I, the identity matrix. Then zero is an eigen
(A) P but not of Q

(B) Q but not of P

(C) both P and Q

(D) neither P nor Q

(C) 1

(D) 0

1 1
is
34. The rank of the matrix
0 0

(A) 4

(B) 2

(IT-04)

35. Consider the following system of linear equations

1 2 3
1 3 4

2 2 3

x 6
y = 8 . The solution of the above system is

z 12

(A) x = 5, y = 3, z = 3

(B) x = 6, y = 4, z = 3

(C) x = 6, y = 3, z = 2

(D) x = 6, y = 3, z = 3

36. Let A, B, C, D be n n matrices, each with non-zero determinant. If ABCD = I,


(CS-04)

then B1 is.
(A)

D1

C1

A1

(C) ADC

(B) CDA
(D) Does not necessarily exist

37. How many solutions does the following system of linear equations have?

(CS-04)

x + 5y = 1 ; x 5 = 2; x + 3y = 3
(A) Infinitely many

(B) Two distinct solutions

(C) Unique

(D) None

38. The following system of linear equations x + 2y = 4; 2x + y = 3; x + y = 1


(A) has unique solution

(B) has no solution

(C) has many solutions

(D) has finite solution

39. In an M N matrix such that all non-zero entries die covered in n rows and b columns. Then
the maximum number of nonzero entries, such that no two are on the same row or column, is:
(CS-04)
(A) a + b

(B) max ( a, b )

(C) min ( M a, N b )

(D) min ( a, b )

You might also like