ITERATIVE

METHODS OF ROOT-FINDING
NUMERICAL METHODS AND ANALYSIS
LESSON 2
PRESENTED BY: ENGR. ROMULO C. CRUZ JR.
ITERATIVE OPEN
METHODS
LESSON 2
 ITERATIVE OPEN METHODS

➢ Fixed-point iteration method

➢ Newton-Raphson method

➢ Secant method

NOTE: Open methods do not always converge

Presented by: Engr. Romulo C. Cruz Jr.

 FIXED-POINT ITERATION

The equation f(x) = 0 can be transformed into x = g(x), and choosing an initial
guess x0, successive approximations can be obtained:
x1 = g(x0)
x2 = g(x1)
⋮
xn+1 = g(xn)

The solution of x = g(x) is called the fixed point of g.

Presented by: Engr. Romulo C. Cruz Jr.

EXAMPLE PROBLEM 5

Find a positive root of the equation x4 + 2x – 8 = 0. Round off to 3 decimal digits.

SOLUTION:
The equation can be written as
8
𝑥 = 𝑥 3 +2  𝑔(𝑥)

By fixed point iteration, this can be generalized as

8
𝑥𝑛+1 = 𝑥 3  xn+1 = g(xn)
𝑛 +2

If 𝑥0 = 1, then
8 8
𝑥1 = 𝑥 3 +2
= 13+2 = 2.667
0
 EXAMPLE PROBLEM 5

8 8
𝑥2 = 𝑥 3 +2 = 2.6673 +2 = 0.382
1
8 8
𝑥3 = 𝑥 3 +2 = 0.3823 +2 = 3.892
2
8 8
𝑥4 = 𝑥 3 = 3.8923 +2 = 0.131
3 +2
8 8
𝑥5 = 𝑥 3 +2 = 0.1313 +2 = 3.995
4
8 8
𝑥6 = = = 0.122
𝑥5 3 +2 3.9953 +2
8 8
𝑥7 = = = 3.996
𝑥6 3 +2 0.1223 +2

The iterative process is diverging (does not converge into a solution)

Presented by: Engr. Romulo C. Cruz Jr.

 EXAMPLE PROBLEM 5

Let us re-write the equation into

1Τ
𝑥 = 8 − 2𝑥 4

or by fixed point iteration

1Τ
𝑥𝑛+1 = 8 − 2𝑥𝑛 4

If 𝑥0 =1, then
1Τ
𝑥1 = 8 − 2(1) 4 = 1.565
1Τ
𝑥2 = 8 − 2(1.565) 4 = 1.486
1Τ
𝑥3 = 8 − 2(1.486) 4 = 1.497

Presented by: Engr. Romulo C. Cruz Jr.

 EXAMPLE PROBLEM 5

1Τ
𝑥4 = 8 − 2(1.497) 4 = 1.496
1Τ
𝑥5 = 8 − 2(1.496) 4 = 1.496

This converges to 1.496. For other values of x0, refer to the Table below:

n 0 1 2 3 4 5 6
𝑥𝑛 1 1.565 1.486 1.498 1.496 1.496
𝑥𝑛 2 1.414 1.508 1.494 1.496 1.496
𝑥𝑛 3 1.189 1.540 1.489 1.497 1.496 1.496

Presented by: Engr. Romulo C. Cruz Jr.

EXAMPLE PROBLEM 6

Find a negative root of the equation x3 + 2x2 – x – 1 = 0. Round off to 3 decimal digits.

SOLUTION:
1Τ
Re-write as 𝑥 = 1 + 𝑥 − 2𝑥 2 3

1Τ
By fixed point iteration: 𝑥𝑛+1 = 1 + 𝑥𝑛 − 2𝑥𝑛 2 3

If 𝑥0 = 1, then
1Τ
𝑥1 = 1 + 1 − 2(1)2 3 =0
1Τ
𝑥2 = 1 + 0 − 2(0)2 3 =1
1Τ
𝑥3 = 1 + 1 − 2(1)2 3 = 0 ==> Diverging!
EXAMPLE PROBLEM 6

If 𝑥0 = 2, then
1Τ
𝑥1 = 1 + 2 − 2(2)2 3 = −1.710
1Τ
𝑥2 = 1 − 1.710 − 2(−1710)2 3 = −1.872
1Τ
𝑥3 = 1 − 1.872 − 2(−1.872)2 3 = −1.990
1Τ
𝑥4 = 1 − 1.990 − 2(−1.990)2 3 = −2.073
1Τ
𝑥5 = 1 − 2.073 − 2(−2.073)2 3 = −2.130

Refer to Table for summary of computed values:

EXAMPLE PROBLEM 6

n 0 1 2 3 4 5 6 7 8 9 10
xn 1 0 1 0
xn 2 -1.710 -1.872 -1.990 -2.073 -2.130 -2.169 -2.195 -2.213 -2.225 -2.232

n 11 12 13 14 15 16 17
xn
xn -2.237 -2.240 -2.242 -2.244 -2.245 -2.246 -2.246

Therefore, the negative root is −2.246

 NEWTON-RAPHSON METHOD

➢ a root-finding method based on

the tangent line to the graph of
the function 𝑦 = 𝑓 𝑥
➢ from Isaac Newton and Joseph
Raphson

Presented by: Engr. Romulo C. Cruz Jr.

 NEWTON-RAPHSON METHOD

The tangent line to f(x) at the point [x0, f(x0)] is

𝑓 𝑥0
𝑡𝑎𝑛𝜃 = 𝑥 = 𝑓 ′ (𝑥0 )
0 −𝑥1
𝑓 𝑥0
𝑥1 = 𝑥0 − 𝑓′ 𝑥0

In a similar manner,
𝑓 𝑥1
𝑥2 = 𝑥1 −
𝑓′ 𝑥1

By considering successive tangent lines, these can be generalized into the

iteration formula
𝑓 𝑥𝑛
𝑥𝑛+1 = 𝑥𝑛 − 𝑓′ 𝑥𝑛

Presented by: Engr. Romulo C. Cruz Jr.

 EXAMPLE PROBLEM 7

Using Newton-Raphson method, find a positive root of the equation x4 + 2x – 8 = 0.

SOLUTION by Newton-Raphson method:

𝑓 𝑥𝑛
𝑥𝑛+1 = 𝑥𝑛 − 𝑓′ 𝑥𝑛
where f(x) = x4 + 2x – 8
f’(x) = 4x3 + 2

If 𝑥0 = 1, then
𝑓 𝑥0 = 14 + 2 1 − 8 = −5
𝑓 ′ 𝑥0 = 4 1 3 +2=6

Presented by: Engr. Romulo C. Cruz Jr.

 EXAMPLE PROBLEM 7

𝑓 𝑥0 −5
𝑥1 = 𝑥0 − 𝑓′ = 1− = 1.833
𝑥0 6
𝑓 𝑥1 6.955
𝑥2 = 𝑥1 − 𝑓′ = 1.833 − = 1.572
𝑥1 26.635
𝑓 𝑥2 1.251
𝑥3 = 𝑥2 − 𝑓′ = 1.572 − = 1.501
𝑥2 17.539
𝑓 𝑥3 0.078
𝑥4 = 𝑥3 − 𝑓′ 𝑥3
= 1.501 − 15.527
= 1.496
𝑓 𝑥4 7.16x10−4
𝑥5 = 𝑥4 − = 1.496 − = 1.496
𝑓′ 𝑥4 15.392

The Table shows the summary of computed values:

Presented by: Engr. Romulo C. Cruz Jr.

 EXAMPLE PROBLEM 7

Iteration n xn f(xn) f’(xn) xn+1

1 0 1 -5 6 1.833
2 1 1.833 6.955 26.635 1.572
3 2 1.572 1.251 17.539 1.501
4 3 1.501 0.078 15.527 1.496
5 4 1.496 7.16x10-4 15.392 1.496

Therefore, as in the Fixed-point iteration method the positive root is 1.496

Presented by: Engr. Romulo C. Cruz Jr.

 LET’S TRY THIS: EXAMPLE PROBLEM 8

Using Newton-Raphson methods, find a positive root of the equation

x sinx + cosx = 0. Round off to 3 decimal digits.

Presented by: Engr. Romulo C. Cruz Jr.

 SECANT METHOD

➢ uses two roots of a secant line

(passing through two points of
a curve) to approximate the
root of a function f(x)

Presented by: Engr. Romulo C. Cruz Jr.

 SECANT METHOD

The slope of the secant line at points (xn+1, f(xn+1)) and (xn, f(xn)) is equal to the
slope at points (xn, f(xn)) and (xn-1,f(xn-1)):

𝑓 𝑥𝑛 −𝑓(𝑥𝑛+1 ) 𝑓 𝑥𝑛−1 −𝑓(𝑥𝑛 )

𝑚= =
𝑥𝑛 −𝑥𝑛+1 𝑥𝑛−1 −𝑥𝑛

Since f(xn+1) = 0, this will yield the iteration formula

𝑓 𝑥𝑛 (𝑥𝑛 −𝑥𝑛−1 )
𝑥𝑛+1 = 𝑥𝑛 −
𝑓 𝑥𝑛 −𝑓(𝑥𝑛−1 )

Presented by: Engr. Romulo C. Cruz Jr.

 SECANT METHOD

Note that it is not advisable to simplify this into

𝑥𝑛−1 𝑓 𝑥𝑛 −𝑥𝑛 𝑓(𝑥𝑛−1 )

𝑥𝑛+1 =
𝑓 𝑥𝑛 −𝑓(𝑥𝑛−1 )

as this may lead to loss of significant digits (if 𝑥𝑛 and 𝑥𝑛−1 are almost equal)

Presented by: Engr. Romulo C. Cruz Jr.

 EXAMPLE PROBLEM 9

Using Secant method, find a positive root of the equation x4 + 2x – 8 = 0. Round off to 3 decimal
digits.

SOLUTION by Secant method:

𝑓 𝑥𝑛 (𝑥𝑛 −𝑥𝑛−1 )
𝑥𝑛+1 = 𝑥𝑛 − 𝑓 𝑥𝑛 −𝑓(𝑥𝑛−1 )

where: 𝑓 𝑥 = 𝑥 4 + 2𝑥 − 8

Let 𝑥0 = 1, 𝑥1 = 2
𝑓(𝑥1 ) 𝑥1 −𝑥0 12 2−1
𝑥2 = 𝑥1 − =2− = 1.294
𝑓 𝑥1 −𝑓(𝑥0 ) 12− −5

Presented by: Engr. Romulo C. Cruz Jr.

 EXAMPLE PROBLEM 9

𝑓(𝑥2 ) 𝑥2 − 𝑥1 −2.608 1.294 − 2

𝑥3 = 𝑥2 − = 1.294 − = 1.420
𝑓 𝑥2 − 𝑓(𝑥1 ) −2.608 − 12

𝑓(𝑥3 ) 𝑥3 − 𝑥2 −1.094 1.420 − 1.294

𝑥4 = 𝑥3 − = 1.420 − = 1.511
𝑓 𝑥3 − 𝑓(𝑥2 ) −1.094 − (−2.608)

𝑓(𝑥4 ) 𝑥4 − 𝑥3 0.235 1.511 − 1.420

𝑥5 = 𝑥4 − = 1.511 − = 1.495
𝑓 𝑥4 − 𝑓(𝑥3 ) 0.235 − (−1.094)

Presented by: Engr. Romulo C. Cruz Jr.

 EXAMPLE PROBLEM 9

𝑓(𝑥5 ) 𝑥5 − 𝑥4 −0.015 1.495 − 1.511

𝑥6 = 𝑥5 − = 1.495 − = 1.496
𝑓 𝑥5 − 𝑓(𝑥4 ) −0.015 − 0.235

𝑓(𝑥6 ) 𝑥6 −𝑥5 7.16x10−4 1.496−1.495

𝑥7 = 𝑥6 − = 1.496 − = 1.496
𝑓 𝑥6 −𝑓(𝑥5 ) 7.16x10−4 −(−0.015)

Presented by: Engr. Romulo C. Cruz Jr.

 EXAMPLE PROBLEM 9

Therefore, as in the Newton-Raphson and fixed-point iteration methods, the root is 1.496.
The Table shows the summary of the calculated values.

Iteration n xn xn-1 f(xn) f(xn-1) xn+1

1 1 2 1 12 -5 1.294

2 2 1.294 2 -2.608 12 1.420

3 3 1.420 1.294 -1.094 -2.608 1.511

4 4 1.511 1.420 0.235 -1.094 1.495

5 5 1.495 1.511 -0.015 0.235 1.496

6 6 1.496 1.495 7.16x10-4 -0.015 1.496

Presented by: Engr. Romulo C. Cruz Jr.

 LET’S TRY THIS: EXAMPLE PROBLEM 10

Using Secant method, find a positive root of the equation x sinx + cosx = 0.
Round off to 3 decimal digits.

SOLUTION:
Iteration n xn xn-1 f(xn) f(xn-1) xn+1
1 1 3 2 - 0.567 1.402 2.712
Let 𝑥0 = 2 , 𝑥1 = 3
2 2 2.712 3 0.220 - 0.567 2.793
𝑓 𝑥 = 𝑥 sin 𝑥 + cos 𝑥
𝑓 𝑥𝑛 (𝑥𝑛 −𝑥𝑛−1 ) 3 3 2.793 2.712 0.014 0.220 2.799
𝑥𝑛+1 = 𝑥𝑛 − 𝑓 𝑥𝑛 −𝑓(𝑥𝑛−1 )
4 4 2.799 2.793 - 0.002 0.014 2.798

5 5 2.798 2.799 - 0.001 - 0.002 2.797

 the positive root is 2.798
6 6 2.797 2.798 0.004 - 0.001 2.798

Presented by: Engr. Romulo C. Cruz Jr.

 LET’S TRY THIS: PRACTICE TASK

Determine a root of the function

f(x) = x3 + x – 1

by: (i) fixed point iteration, (ii) Newton-Raphson method, and (iii)
secant method. Round off your answer to three decimal places.

Presented by: Engr. Romulo C. Cruz Jr.

 https://www.clipartkey.com/view/miowRw_thank-you-thank-you-in-different-languages-png/
Presented by: Engr. Romulo C. Cruz Jr.