0% found this document useful (0 votes)
107 views6 pages

Newton-Raphson Method Explained

The document summarizes using the bisection method and Newton's method to find the real roots of two polynomial equations. For the bisection method, it takes 49 steps to find a root of x^3 + 4x^2 - 10 to machine precision. For Newton's method, it takes 9 steps to find a root of x^3 + x - 1 to machine precision, starting from an initial guess of -2.4. Diagrams and calculations are shown to illustrate the convergence of each method to the solution.

Uploaded by

fake7083
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
107 views6 pages

Newton-Raphson Method Explained

The document summarizes using the bisection method and Newton's method to find the real roots of two polynomial equations. For the bisection method, it takes 49 steps to find a root of x^3 + 4x^2 - 10 to machine precision. For Newton's method, it takes 9 steps to find a root of x^3 + x - 1 to machine precision, starting from an initial guess of -2.4. Diagrams and calculations are shown to illustrate the convergence of each method to the solution.

Uploaded by

fake7083
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

bisection method (x^3) + (4*x^2) -10 =0

Input interpretation

Input interpretation:

solve x3 + 4 x2 - 10 0 using bisection method to machine precision

Result:

x 1.365230013414097
(using starting points of a0 1. and b0 2.)

Symbolic form of bisection method:


1
cn 2
(an + bn )
[cn , bn ] sgn(f (cn )) sgn(f (an ))
[an+1 , bn+1 ]
[an , cn ] (otherwise)
where f (x) x3 + 4 x2 - 10

sgn(x) is the sign of x

Steps: More

49 steps to machine precision

Convergence diagram:

30

20
b0

10
b1

1.0 b49 1.5 2.0 2.5


a2
a0

Interval convergence diagram:

45
40
35
30
step

25
20
15
10
5
0
2

0
1.0 1.2 1.4 1.6 1.8 2.0
x

Convergence rates:

1
10-20
machine precision
residual

10-40
10-60 precision 50
10-80
10-100 precision 100
1 34 67 100133166199232265298331
step
(using starting points of a0 1. and b0 2.)

precision steps

machine precision 49

50 digits 164

100 digits 331

Exact result: More digits

Solve[-10 + 4 * x ^ 2 + x ^ 3 == 0, {x}]

1 3 3
x -4 + 71 - 3 105 + 71 + 3 105
3
1.365230013414096845760806828981666078331

30

20
b0

Out[3]=
10
b1

1.0 b49 1.5 2.0 2.5


a2
a0

In[5]:= Newton-Raphson Method (x^3)+x-1=0


Result

Input interpretation:

solve x3 + x - 1 0 using Newton's method to machine precision


3

Result:

FindRoot[-1 + x + x ^ 3, {x, -2.4},


Method -> {"Newton", "StepControl" -> None}]

x 0.6823278038280193
(using starting point of x0 -2.4)

Symbolic form of Newton iteration: Hide details

x3n +xn -1
xn+1 xn - 3 x2n +1
f (xn )
xn+1 xn - f (x )
n

where f (x) x3 + x - 1
f (x) 3 x2 + 1

Steps: More Hide details

9 steps to machine precision

Less

x0 -2.4
x30 +x0 -1
x1 x 0 - 3 x20 +1
x1 -2.4 - (-0.942232)
x1 -1.45777
x31 +x1 -1
x2 x 1 - 3 x21 +1
x2 -1.45777 - (-0.753282)
x2 -0.704486
x32 +x2 -1
x3 x 2 - 3 x22 +1
x3 -0.704486 - (-0.825313)
x3 0.120827
x33 +x3 -1
x4 x 3 - 3 x23 +1
x4 0.120827 - (-0.840593)
x4 0.96142
x34 +x4 -1
x5 x 4 - 3 x24 +1
x5 0.96142 - (0.225309)
x5 0.736111
x35 +x5 -1
x6 x 5 - 3 x25 +1
x6 0.736111 - (0.0514095)
x6 0.684701
x36 +x6 -1
x7 x 6 - 3 x26 +1
x7 0.684701 - (0.00236889)
x7 0.682333
x37 +x7 -1
x8 x 7 - 3 x27 +1
x 0.682333 (4.8038710-6 )
4

x8 0.682333 - (4.8038710 6 )
x8 0.682328
x38 +x8 -1
x9 x 8 - 3 x28 +1
x9 0.682328 - (1.9709910-11 )
x9 0.682328

Iteration diagram:

20

-4 -3 -2 -1 1 2

-20

-40

-60

Convergence by starting point diagram:

12
10
steps

8
6
x
x0
-10 -5 0 5 10

starting point
(color indicates final value, x , converged to)

Convergence rates:

1
10-20
machine precision
residual

10-40
10-60 precision 50
10-80
10-100 precision 100
1 2 3 4 5 6 7 8 9 10 11 12 13

step
(using starting point of x0 -2.4)

precision steps

machine precision 9

50 digits 11
5

100 digits 12

Exact result: More digits

Solve[-1 + x + x ^ 3 == 0, {x}]

1
3
2
9 + 93
2
x- +
3
3 9 + 93 32/3

0.6823278038280193273694837397110482568912

Higher-order Newton iterations: Less


3rd-order iteration (Householder's iteration):
7 5 4 3
xn+1 21 xn +18 xn +32xn +53 xn +3 xn +1
3 xn +1
f (xn )2 f (xn )
xn+1 - ff(x(xn )) - 2 f (xn )3
+ xn
n

where f (x) x3 + x - 1
f (x) 3 x2 + 1
f (x) 6 x

step x

0 -2.400000000000000

1 -1.10809

2 0.0218168

3 0.936238

4 0.694057

5 0.682329

6 0.682328

7 0.682328

4th order iteration:


11 9 8 7 6 5 4 3 2
xn+1 204 xn +332 xn -18 xn +216 xn -69
2
xn +132 xn -27 xn +64 xn -6 xn +2
5
3 xn +1
2
f (xn )3 3ff (x(xn )n4) - ff (x(xn )n3)
(3)

f (xn )2 f (xn )
xn+1 - ff(x(xn )) - 2 f (xn )3
- 6 f (x
+ xn
n n)

where f (x) + x - 1 x3
f (x) 3 x2 + 1
f (x) 6 x
f (3) (x) 6

step x
6

0 -2.400000000000000

1 -0.894301

2 0.768988

3 0.682393

4 0.682328

5 0.682328

Mathematica input:

FindRoot[-1 + x + x ^ 3, {x, -2.4},


Method -> {"Newton", "StepControl" -> None}]

FindRoot[-1 + x + x ^ 3, {x, -2.4},


Method-> {"Newton", "StepControl" -> None}]

Out[5]=
x 0.6823278038280193
(using starting point of x0 -2.4)

You might also like