Instructor

Majd Shhadi, PhD

Fall 2023
1
~ Truncation Errors and the Taylor Series ~

The slides’ copyright is to

Eng. Muhannad Al-Jabi & Dr. Aysar Yasin
 What is a Taylor series?
3

 Some examples of Taylor series which you must have

seen
x2 x4 x6
cos( x) = 1 − + − +
2! 4! 6!

x3 x5 x7
sin( x) = x − + − +
3! 5! 7!

x2 x3
e = 1+ x +
x
+ +
2! 3!

Majd Shhadi, PhD

 What is a Taylor series?
4

x2 x3
e = 1+ x +
x
+ +
2! 3!

The exponential function ex (in blue),

and the sum of the first n + 1 terms
of its Taylor series at 0 (in red).

Majd Shhadi, PhD

 General Taylor Series
5

 The general form of the Taylor series is given by

𝑓 ″ 𝑥 𝑓 ‴ 𝑥
𝑓 𝑥 + ℎ = 𝑓 𝑥 + 𝑓′ 𝑥 ℎ + ℎ2 + ℎ3 + ⋯
2! 3!

provided that all derivatives of f(x) are continuous and exist in

the interval [x, x+h]
 In mathematics, a Taylor series is a representation of a
function as an infinite sum of terms that are calculated from
the values of the function’s derivatives at a single point.
 If the Taylor series is centered at zero, then that series is also
called a Maclaurin series.
 4.1: The Taylor series
6

Taylor’s Theorem
If the function f and its first n+1 derivatives are continuous on an
interval containing 𝑥𝑖 and 𝑥𝑖+1 , then the value of the function at 𝑥𝑖+1
is given by:

where the remainder Rn is defined as

 is some point in the domain [𝒙𝒊 , 𝒙𝒊+𝟏 ]

 4.1: The Taylor series
7

Simple form
𝒉 = 𝒙𝒊+𝟏 − 𝒙𝒊 h: step size

A remainder term is included to account for all terms from n+1 to

infinity

 is some point in the domain [𝒙𝒊 , 𝒙𝒊+𝟏 ]

 4.1: The Taylor series
8

Graphical depiction of a zero-order Taylor series prediction and

remainder.
 The Taylor series
9

Using Taylor’s Theorem, we can approximate any

smooth function by a polynomial.

The zero-order approximation of the value of f(x)

is given 𝑓 𝑥𝑖+1 ≈ 𝑓(𝑥𝑖 )

The first-order approximation is given by

𝑓 𝑥𝑖+1 ≈ 𝑓 𝑥𝑖 + 𝑓ሗ (𝑥𝑖 )h

Eng. Muhannad Al-Jabi

 Example-Taylor Series
10

 Find the value of 𝒇 𝟔 given that

𝑓 4 = 125, 𝑓 ′ 4 = 74, 𝑓 ″ 4 = 30, 𝑓 ‴ 4 = 6
and all other higher order derivatives of 𝑓 𝑥 at 𝑥 = 4 are
zero
 Solution
ℎ 2 ℎ 3
𝑓 𝑥 + ℎ = 𝑓 𝑥 + 𝑓′ 𝑥 ℎ + 𝑓″ 𝑥 + 𝑓‴ 𝑥 +⋯
2! 3!

𝑥=4

ℎ =6−4=2
 Example (cont.)
11

Solution: (cont.)
Since the higher order derivatives are zero,
ℎ 2 ℎ 3
𝑓 𝑥 + ℎ = 𝑓 𝑥 + 𝑓′ 𝑥 ℎ + 𝑓″ 𝑥 + 𝑓‴ 𝑥 +⋯
2! 3!
2 3
2 2
𝑓 4 + 2 = 𝑓 4 + 𝑓′ 4 2 + 𝑓″ 4 + 𝑓‴ 4
2! 3!

= 125 + 148 + 60 + 8 = 341

Note that to find 𝑓 6 exactly, we only need the value of the

function and all its derivatives at some other point, in this case 𝒙 = 𝟒
 Example 4.2
12

 Use Taylor series expansions with n = 0 to 6 to approximate

f(x) = cos x at xi+1= /3 on the basis of the value of f(x) and
its derivatives at xi = /4.
 Solution
𝜋
h = /3 - /4 = /12 and the true value 𝑓 = 0.5
3

Zero-order approximation: f(/3)  cos (/4) = 0.707106781

0.5 − 0.707106781
𝜀𝑡 = 100% = 41.4%
0.5

First-order approximation: f(/3)  cos (/4) – (/12) sin (/4)

= 0.521986659
Eng. Muhannad Al-Jabi
 Example 4.2: Solution
13

Order f(n)(x) Result εt%

0 Cos(x) 0.707106781 41.42
1 -Sin(x) 0.521986659 4.41
2 -Cos(x) 0.497754491 0.45
3 Sin(x) 0.499869147 0.026
4 Cos(x) 0.500007551 0.0015
5 -Sin(x) 0.500000304 6.08E-05
6 -Cos(x) 0.499999988 2.44E-06

𝝅
True value of 𝒇 = 𝟎. 𝟓
𝟑
Eng. Muhannad Al-Jabi
 Notes
14

To get more accurate estimation of f(x), we can do one or

both of the following:

 Add more terms to the Taylor polynomial

 Reduce the value of h.

Eng. Muhannad Al-Jabi

 Example
15

Use zero-through fourth order Taylor series expansions

for f(x)=ln(x) using the base point at xi =1 to predict
the value of f(x)
1
a) At xi+1= 2.5 f ' ( x) =
x
b) At xi+1 =1.5 1
f ' ' ( x) = −
x2
2
f ' ' ' ( x) =
x3
6
f ' ' ' ' ( x) = −
x4
 Example: Solution a
16

True value for ln(2.5) =0.916291

# of order Result εt%
0 0 100
1 1.5 63.70345
2 0.375 59.07414
3 1.5 63.70345
4 0.234375 74.42134
5 ?? ??
 Example: Solution b
17

True value for ln(1.5) =0.405465

# of order Result εt
0 0 100
1 0.5 23.31521
2 0.375 7.513596
3 0.416667 2.762672
4 0.401042 1.090929
5 ?? ??

We can see that as the step size (h) decreased the Taylor
series becomes more accurate
 Example 4.3
Suppose that you have the following function

f ( x) = x m

Use zero-through fourth order Taylor

series expansions to predict 𝒇(𝟐)
using the base point at xi=1

a) When m=1
b) When m=2
c) When m=3
d) When m=4
 Example 4.3: Solution m=1
19

f ( x) = x1

ℎ 2 ℎ 3
𝑓 𝑥𝑖+1 = 𝑓 𝑥𝑖 + 𝑓 ′ 𝑥𝑖 ℎ + 𝑓 ″ 𝑥 + 𝑓 ‴ 𝑥𝑖 +⋯
2! 3!

 Zero order 𝒇(𝒏+𝟏) (𝝃)

𝑹𝒏 = (𝒉)𝒏+𝟏
𝑓 𝑥𝑖+1 = 𝑓 𝑥𝑖 (𝒏 + 𝟏)!

𝑹𝟎 = 𝒇′(𝝃)(𝒉)
𝑓(2) = 𝑓(1) = 1

𝑹𝟎 = 𝒇′ 𝟏 ∗ 𝟏 = 𝟏
 Example 4.3: Solution m=1
20

f ( x) = x1
ℎ 2 ℎ 3
𝑓 𝑥𝑖+1 = 𝑓 𝑥𝑖 + 𝑓 ′ 𝑥𝑖 ℎ + 𝑓 ″ 𝑥 + 𝑓 ‴ 𝑥𝑖 +⋯
2! 3!
 First order
𝒇(𝒏+𝟏) (𝝃)
𝑹𝒏 = (𝒉)𝒏+𝟏
𝑓 𝑥𝑖+1 = 𝑓 𝑥𝑖 + 𝑓 ′ 𝑥𝑖 ℎ (𝒏 + 𝟏)!

𝒇(𝟐) (𝝃)
𝑹𝟏 = (𝒉)𝟐
𝑓(2) = 𝑓(1) + 𝑓′(1) ∗ 1 = 2 (𝟐)!

𝒇𝟐 𝟏
𝑹𝟏 = (𝟏)𝟐 = 𝟎
𝟐 ∗𝟏
 Example 4.3: Solution m=1
21

f ( x) = x1
ℎ 2 ℎ 3
𝑓 𝑥𝑖+1 = 𝑓 𝑥𝑖 + 𝑓 ′ 𝑥𝑖 ℎ + 𝑓 ″ 𝑥 + 𝑓 ‴ 𝑥𝑖 +⋯
2! 3!
 Second order
ℎ 2 𝒇(𝒏+𝟏) (𝝃)
𝑓 𝑥𝑖+1 = 𝑓 𝑥𝑖 + 𝑓 ′ 𝑥𝑖 ℎ + 𝑓 ″ 𝑥 𝑹𝒏 = (𝒉)𝒏+𝟏
2! (𝒏 + 𝟏)!

𝒇𝟑 𝟏
𝑓 2 = 𝑓 1 + 𝑓′ 1 ∗ 1 + 0 = 2 𝑹𝟐 = (𝟏)𝟑 = 𝟎
𝟑∗𝟐 ∗𝟏

So , the result of third and fourth order will be 2.

 Example 4.3: Solution m=2
22

𝑓(𝑥) = 𝑥 2

ℎ 2 ℎ 3
𝑓 𝑥𝑖+1 = 𝑓 𝑥𝑖 + 𝑓 ′ 𝑥𝑖 ℎ + 𝑓 ″ 𝑥 + 𝑓 ‴ 𝑥𝑖 +⋯
2! 3!

 Zero order 𝒇(𝒏+𝟏) (𝝃)

𝑹𝒏 = (𝒉)𝒏+𝟏
𝑓 𝑥𝑖+1 = 𝑓 𝑥𝑖 (𝒏 + 𝟏)!
𝑓(2) = 𝑓(1) = 1
𝑹𝟎 = 𝒇′(𝝃)(𝒉)

𝑹𝟎 = 𝒇′ 𝟏 ∗ 𝟏 =2
 Example 4.3: Solution m=2
23

𝑓(𝑥) = 𝑥 2
ℎ 2 ℎ 3
𝑓 𝑥𝑖+1 = 𝑓 𝑥𝑖 + 𝑓 ′ 𝑥𝑖 ℎ + 𝑓 ″ 𝑥 + 𝑓 ‴ 𝑥𝑖 +⋯
2! 3!
 First order
𝒇(𝒏+𝟏) (𝝃)
𝑹𝒏 = (𝒉)𝒏+𝟏
𝑓 𝑥𝑖+1 = 𝑓 𝑥𝑖 + 𝑓 ′ 𝑥𝑖 ℎ (𝒏 + 𝟏)!

𝒇(𝟐) (𝝃)
𝑹𝟏 = (𝒉)𝟐
𝑓 2 = 𝑓 1 + 𝑓′ 1 ∗ 1 (𝟐)!
= 1+2 =3
𝒇𝟐 𝟏
𝑹𝟏 = (𝟏)𝟐 = 𝟏
𝟐 ∗𝟏
 Example 4.3: Solution m=2
24

𝑓(𝑥) = 𝑥 2

ℎ 2 ℎ 3
𝑓 𝑥𝑖+1 = 𝑓 𝑥𝑖 + 𝑓 ′ 𝑥𝑖 ℎ + 𝑓 ″ 𝑥 + 𝑓 ‴ 𝑥𝑖 +⋯
2! 3!
 Second order
ℎ 2 𝒇(𝒏+𝟏) (𝝃)
𝑓 𝑥𝑖+1 = 𝑓 𝑥𝑖 + 𝑓 ′ 𝑥𝑖 ℎ + 𝑓 ″ 𝑥 𝑹𝒏 = (𝒉)𝒏+𝟏
2! (𝒏 + 𝟏)!

1
𝑓 2 = 𝑓 1 + 𝑓′ 1 ∗ 1 + 𝑓″ 1 𝒇𝟑 𝟏
2 𝑹𝟐 = (𝟏)𝟑 = 𝟎
= 1+2(1)+2(0.5) = 4 𝟑∗𝟐 ∗𝟏

So, the second, third and fourth order will give us a result = 4
 Example 4.3: Solution m=3
25

f ( x) = x 3

f (2) = f (1) = 1
 Zero order
R0 = f ' (1) *1 = 3 *12 = 3

 First order f (2) = f (1) + f ' (1) *1 = 4

f ' ' (1) *12 6
R1 = = =3
2 2

 Second order f (2) = f (1) + f ' (1) *1 +

f ' ' (1) *12
=7
2
f ' ' ' (1) *13 6
R2 = = =1
3! 6
 Example 4.3: Solution m=3
26

 Third order
f ' ' (1) *12 f ' ' ' (1) *13
f (2) = f (1) + f ' (1) *1 + + =8
2 3!
f ' ' ' ' (1) *13 0
R3 = = =0
4! 24

So, the third and fourth order will give us a result =8

So, when m=4 the fourth order and higher will give us the
true value which is 16
 Example
27

 Use first-through third-order Taylor series to predict

𝑓(1.6) for:
𝒇 𝒙 = 𝒔𝒊𝒏𝒙 + 𝒄𝒐𝒔 𝟏 + 𝒙𝟐 − 𝟏

 Use base point at x=1. Compute the true percent

relative error ɛt for each approximation.

Majd Shhadi, PhD

 Example: Solution
28

 Use first-through third-order Taylor series to predict

𝑓(1.6) for:
𝒇 𝒙 = 𝒔𝒊𝒏(𝒙) + 𝒄𝒐𝒔 𝟏 + 𝒙𝟐 − 𝟏

 Use base point at x=1. Compute the true percent

relative error ɛt for each approximation.

ℎ = 1.6 − 1 = 0.6
( 3)
f ' ' (1) f (1)
f (1.6 ) = f (1) + f ' (1)(0.6) + ( 0. 6 ) +
2
( 0. 6 ) 3
2! 3!
Majd Shhadi, PhD
 Example: Solution
29

True value f(1.6)= -0.91416

Order f(n)(x) f(n)(1) Sum εt%

0 𝑠𝑖𝑛(𝑥) + 𝑐𝑜𝑠 1 + 𝑥 2 − 1 -0.57468 -0.57468 37.13

1 cos(𝑥) − 2𝑥𝑠𝑖𝑛 1 + 𝑥 2 -1.2782 -1.3417 46.75

-𝑠𝑖𝑛(𝑥) − 4𝑥 2 𝑐𝑜𝑠 1 + 𝑥 2
2 -0.99548 -1.5208 66.34
− 2𝑠𝑖𝑛 1 + 𝑥 2
-cos 𝑥 − 12𝑥𝑐𝑜𝑠 1 + 𝑥 2
11.7278 -1.0986 20.17
3 + 8𝑥 3 𝑠𝑖𝑛 1 + 𝑥 2

Majd Shhadi, PhD

 Example
30

 Use first to third order Taylor series expansion to find 𝑓(1.5)

with 𝑥0 = 1.2 as a base point for the following function:

𝒇 𝒙 = 𝒄𝒐𝒔 𝟏 + 𝒙𝟐 + 𝒍𝒐𝒈(𝒙)

Compute the true relative percent error ɛt after each case. Carry
five significant figures during the computation (Show your
complete solution).

Majd Shhadi, PhD

 Example: Solution
31

′ 2
1
𝑓 𝑥 = −2𝑥𝑠𝑖𝑛 1 + 𝑥 +
𝑥
1
𝑓 ′′ 𝑥 = −4𝑥 2 𝑠𝑖𝑛 1 + 𝑥2 − 2𝑠𝑖𝑛 1 + 𝑥2
− 2
𝑥
2
𝑓 ′′′ 𝑥 = 8𝑥 3 𝑠𝑖𝑛 1 + 𝑥 2 − 12𝑥𝑐𝑜𝑠 1 + 𝑥 2 + 3
𝑥

First: 𝒇 𝟏. 𝟓 = 𝒇 𝟏. 𝟐 + 𝒇′ 𝟏. 𝟐 𝒉
𝒉 𝟐
Second: 𝒇 𝟏. 𝟓 = 𝒇 𝟏. 𝟐 + 𝒇′ 𝟏. 𝟐 𝒉 + 𝒇″ (𝟏. 𝟐)
𝟐!
′ ″ 𝒉𝟐 ‴ 𝒉𝟑
Third: 𝒇 𝟏. 𝟓 = 𝒇 𝟏. 𝟐 + 𝒇 𝟏. 𝟐 𝒉 + 𝒇 (𝟏. 𝟐) +𝒇 (𝟏. 𝟐)
𝟐! 𝟑!
h =1.5-1.2= 0.3
Majd Shhadi, PhD
 Example: Solution
32

True value f(1.5)= -0.818038

Order xi fn(xi) f(xi+1) εt%

0 1.2 -0.68463 -0.68463 16.31

1 1.2 -0.71571 -0.89935 9.94

2 1.2 2.41426 -0.79071 3.34

3 1.2 21.07884 -0.69585 14.94

Majd Shhadi, PhD