FINDING EIGENVALUES AND EIGENVECTORS

EXAMPLE 1: Find the eigenvalues and eigenvectors of the matrix

 
1 −3 3
A =  3 −5 3  .
6 −6 4

SOLUTION:

• In such problems, we first find the eigenvalues of the matrix.

FINDING EIGENVALUES

• To do this, we find the values of λ which satisfy the characteristic equation of the
matrix A, namely those values of λ for which

det(A − λI) = 0,

where I is the 3×3 identity matrix.

• Form the matrix A − λI:

     
1 −3 3 λ 0 0 1−λ −3 3
A − λI =  3 −5 3  −  0 λ 0  =  3 −5 − λ 3 .
6 −6 4 0 0 λ 6 −6 4−λ

Notice that this matrix is just equal to A with λ subtracted from each
entry on the main diagonal.
• Calculate det(A − λI):
     
 −5 − λ 3  − (−3)  3 3 
   3 −5 − λ 
det(A − λI) = (1 − λ)  + 3 
−6 4−λ   6 4−λ  6 −6 
= (1 − λ) ((−5 − λ)(4 − λ) − (3)(−6)) + 3(3(4 − λ) − 3 × 6) + 3(3 × (−6) − (−5 − λ)6)
= (1 − λ)(−20 + 5λ − 4λ + λ2 + 18) + 3(12 − 3λ − 18) + 3(−18 + 30 + 6λ)
= (1 − λ)(−2 + λ + λ2 ) + 3(−6 − 3λ) + 3(12 + 6λ)
= −2 + λ + λ2 + 2λ − λ2 − λ3 − 18 − 9λ + 36 + 18λ
= 16 + 12λ − λ3 .

• Therefore
det(A − λI) = −λ3 + 12λ + 16.
REQUIRED: To find solutions to det(A − λI) = 0 i.e., to solve

λ3 − 12λ − 16 = 0. (1)

* Look for integer valued solutions.

1
 * Such solutions divide the constant term (-16). The list of possible integer
solutions is
±1, ±2, ±4, ±8, ±16.

* Taking λ = 4, we find that 43 − 12.4 − 16 = 0.

* Now factor out λ − 4:

(λ − 4)(λ2 + 4λ + 4) = λ3 − 12λ2 + 16.

* Solving λ2 + 4λ + 4 by formula1 gives

√
−4 ± 42 − 4.1.4 −4 ± 0
λ= = ,
2 2
and so λ = −2 (a repeated root).

• Therefore, the eigenvalues of A are λ = 4, −2. (λ = −2 is a repeated root of the

characteristic equation.)

FINDING EIGENVECTORS

• Once the eigenvalues of a matrix (A) have been found, we can find the eigenvectors
by Gaussian Elimination.

• STEP 1: For each eigenvalue λ, we have

(A − λI)x = 0,

where x is the eigenvector associated with eigenvalue λ.

• STEP 2: Find x by Gaussian elimination. That is, convert the augmented matrix

..
 
A − λI . 0

to row echelon form, and solve the resulting linear system by back substitution.

We find the eigenvectors associated with each of the eigenvalues

• Case 1: λ = 4

– We must find vectors x which satisfy (A − λI)x = 0.

1
To find the roots of a quadratic equation of the form ax2 + bx√+ c = 0 (with a√ 6= 0) first compute
∆ = b2 − 4ac, then if ∆ ≥ 0 the roots exist and are equal to x = −b−
2a
∆
and x = −b+
2a
∆
.

2
 – First, form the matrix A − 4I:
 
−3 −3 3
A − 4I =  3 −9 3  .
6 −6 0

..
 
– Construct the augmented matrix A − λI . 0 and convert it to row echelon
form
   
−3 −3 3 0 R1 1 1 −1 0 R1
R1→−1/3×R3
 3 −9 3 0  R2 −→  3 −9 3 0  R2
6 −6 0 0 R3 6 −6 0 0 R3
 
R2→R2−3×R1 1 1 −1 0 R1
R3→R3−6×R1
−→  0 −12 6 0  R2
0 −12 6 0 R3
 
1 1 −1 0 R1
R2→−1/12×R2
−→  0 1 −1/2 0  R2
0 −12 6 0 R3
 
1 1 −1 0 R1
R3→R3+12×R2
−→  0 1 −1/2 0  R2
0 0 0 0 R3
 
1 0 −1/2 0 R1
R1→R1−R2
−→  0 1 −1/2 0  R2
0 0 0 0 R3

– Rewriting this augmented matrix as a linear system gives

x1 − 1/2x3 = 0
x2 − 1/2x3 = 0

So the eigenvector x is given by:

 x3   1

x1 = 2 2
x3 1
x =  x2 = 2
 = x3 
2

x3 1

For any real number x3 6= 0. Those are the eigenvectors of A associated

with the eigenvalue λ = 4.

• Case 2: λ = −2

– We seek vectors x for which (A − λI)x = 0.

– Form the matrix A − (−2)I = A + 2I
 
3 −3 3
A + 2I =  3 −3 3  .
6 −6 6

3
 ..
 
– Now we construct the augmented matrix A − λI . 0 and convert it to row
echelon form
   
3 −3 3 0 R1 1 −1 1 0 R1
R1→1/3×R3
 3 −3 3 0  R2 −→  3 −3 3 0  R2
6 −6 6 0 R3 6 −6 6 0 R3
 
R2→R2−3×R1 1 1 −1 0 R1
R3→R3−6×R1
−→  0 0 0 0  R2
0 0 0 0 R3

– When this augmented matrix is rewritten as a linear system, we obtain

x1 + x2 − x3 = 0,

so the eigenvectors x associated with the eigenvalue λ = −2 are given by:

 
x1 = x3 − x2
x= x2 
x3

– Thus
     
x3 − x2 1 −1
x= x2  = x3  0  + x2  1  for any x2 , x3 ∈ R\{0}
x3 1 0

are the eigenvectors of A associated with the eigenvalue λ = −2.