4.
8 Newton’s method
We wish to find an approximate solution to f (x) = 0.
r: the root that we are trying to find
x1 : an initial guess (a first approximation)
The equation of the tangent line to y = f (x) at x = x1 is
y = f (x1 ) + f ′ (x1 )(x − x1 )
We denote the x-intercept of the tangent line by x2
0 = f (x1 ) + f ′ (x1 )(x2 − x1 )
f (x1 )
⇒ x2 = x1 − if f ′ (x1 ) ̸= 0
f ′ (x1 )
We use x2 as a second approximation to r.
1
Next we repeat this procedure with x1 replaced by x2 using the tangent line at
(x2 , f (x2 )). This give a third approximation:
f (x2 )
x3 = x2 − if f ′ (x2 ) ̸= 0
f ′ (x2 )
In this way, we generate a sequence of successive approximations determined by
f (xn )
xn+1 = xn − , if f ′ (xn ) ̸= 0, n = 1, 2, · · ·
f ′ (xn )
This procedure is called the Newton’s method.
2
Example
Starting with x1 = 2, find the third approximation x3 to the root of the
equation x3 − 2x − 5 = 0.
Solution. Let f (x) = x3 − 2x − 5. Then f ′ (x) = 3x2 − 2.
x3n − 2xn − 5
xn+1 = xn − , n = 1, 2, 3, · · ·
3x2n − 2
With n = 1, we have
x31 − 2x1 − 5
x2 = x1 −
3x21 − 2
23 − 2(2) − 5
=2−
3(2)2 − 2
= 2.1
With n = 2, we obtain
x32 − 2x2 − 5
x3 = x2 −
3x22 − 2
2.13 − 2(2.1) − 5
= 2.1 −
3(2.1)2 − 2
≈ 2.0946
3
Remark
In certain circumstances, this sequence may not converge.
4
Example
√
6
Use Newton’s method to approximate 2.
Solution. Let f (x) = x6 − 2. Then f ′ (x) = 6x5 .
x6n − 2
xn+1 = xn −
6x5n
If we choose x1 = 1 (as the initial approximation), then
x2 ≈ 1.16666667
x3 ≈ 1.12644368
x4 ≈ 1.12249707
x5 ≈ 1.12246205
x6 ≈ 1.12246205
Therefore √
6
2 ≈ 1.12246205.