Numerical Methods in Engineering
ENGG222, Spring 2025
Chapter 4
Truncation Errors and the Taylor Series
Ramin Masoudi
Associate Professor of the Department of Mechanical Engineering
American University in Dubai
ENGG222: Numerical Methods in Engineering Dr. Ramin Masoudi, Mechanical Engineering 1
INTRODUCTION
Truncation errors are those that result from using an approximation in
place of an exact mathematical procedure, for example
Non-elementary functions such as trigonometric, exponential, and
others are expressed in an approximate fashion using Taylor series
when their values, derivatives, and integrals are computed
The Taylor series provides a means to predict a function value at one
point in terms of the function value and its derivatives at another point
ENGG222: Numerical Methods in Engineering Dr. Ramin Masoudi, Mechanical Engineering 2
THE TAYLOR SERIES
The zero-order approximation gives a perfect estimate for a
constant function
The first-order approximation gives a perfect estimate for
a straight trend
The second-order approximation capture some of the
curvature that the function might exhibit
ENGG222: Numerical Methods in Engineering Dr. Ramin Masoudi, Mechanical Engineering 3
The complete Taylor series expansion
𝑥𝑥𝑖𝑖 ≤ 𝜉𝜉 ≤ 𝑥𝑥𝑖𝑖+1
The Remainder term, 𝑅𝑅𝑛𝑛 , accounts for all terms from (𝑛𝑛 + 1) to infinity
ℎ = 𝑥𝑥𝑖𝑖+1 − 𝑥𝑥𝑖𝑖 Step size
ENGG222: Numerical Methods in Engineering Dr. Ramin Masoudi, Mechanical Engineering 4
This remainder relationship, 𝑅𝑅𝑛𝑛 , has two major drawbacks:
First, 𝜉𝜉 is not known exactly but merely lies somewhere between 𝑥𝑥𝑖𝑖 and 𝑥𝑥𝑖𝑖+1
Second, to evaluate remainder equation 𝑅𝑅𝑛𝑛 , we need to determine the
(𝑛𝑛 + 1)th derivative of 𝑓𝑓(𝑥𝑥). To do this, we need to know 𝑓𝑓(𝑥𝑥). However, if we
knew 𝑓𝑓(𝑥𝑥), there would be no need to perform the Taylor series expansion in
the present context!
Despite this dilemma, the remainder relation is still useful for
gaining insight into truncation errors. This is because we do
have control over the term ℎ in the equation
where the nomenclature 𝑂𝑂(ℎ𝑛𝑛+1 ) means that
the truncation error is of the order of ℎ𝑛𝑛+1 .
ENGG222: Numerical Methods in Engineering Dr. Ramin Masoudi, Mechanical Engineering 5
Sample Problem: Use zero- through fourth-order Taylor series expansions to approximate
the function from 𝑥𝑥𝑖𝑖 = 0 with ℎ = 1
In general, the 𝑛𝑛th −order Taylor series expansion
will be exact for an 𝑛𝑛th −order polynomial
ENGG222: Numerical Methods in Engineering Dr. Ramin Masoudi, Mechanical Engineering 6
Sample Problem:
ENGG222: Numerical Methods in Engineering Dr. Ramin Masoudi, Mechanical Engineering 7
The Remainder for the Taylor Series Expansion
Suppose that we truncated the Taylor series expansion after the zero-order term to yield
It is obviously inconvenient to deal with the
remainder in this infinite series format. One
simplification might be to truncate the remainder
itself
ENGG222: Numerical Methods in Engineering Dr. Ramin Masoudi, Mechanical Engineering 8
The derivative mean-value theorem:
If a function 𝑓𝑓(𝑥𝑥) and its first derivative are
continuous over an interval from 𝑥𝑥𝑖𝑖 to 𝑥𝑥𝑖𝑖+1 ,
then there exists at least one point on the
function that has a slope, designated by
𝑓𝑓 ′ (𝜉𝜉), that is parallel to the line joining
𝑓𝑓(𝑥𝑥𝑖𝑖 ) and 𝑓𝑓(𝑥𝑥𝑖𝑖+1 )
Using the same logic:
…….
ENGG222: Numerical Methods in Engineering Dr. Ramin Masoudi, Mechanical Engineering 9
Using the Taylor Series to Estimate Truncation Errors
For the jumper problem, the goal was to estimate velocity
let us truncate the series after the first derivative term
On the other hand, from the previous slide
ENGG222: Numerical Methods in Engineering Dr. Ramin Masoudi, Mechanical Engineering 10
Numerical Differentiation
Finite divided difference is a method of numerical differentiation in numerical methods,
based on the Taylor series expansion
Forward Finite-divided Difference
Approximation
first forward difference
Step size
ENGG222: Numerical Methods in Engineering Dr. Ramin Masoudi, Mechanical Engineering 11
Backward Finite-divided Difference
Approximation
ENGG222: Numerical Methods in Engineering Dr. Ramin Masoudi, Mechanical Engineering 12
Centered Finite-divided Difference
Approximation
From the forward Taylor series expansion:
Substituting 𝑓𝑓(𝑥𝑥𝑖𝑖 ) from the Taylor formulation in
the backward finite-divided difference mentioned in
the previous slide, we have
ENGG222: Numerical Methods in Engineering Dr. Ramin Masoudi, Mechanical Engineering 13
Sample Problem: Use forward and backward difference approximations of 𝑂𝑂(ℎ) and a
centered difference approximation of 𝑂𝑂(ℎ2 ) to estimate the first derivative of
𝑓𝑓(𝑥𝑥) = −0.1𝑥𝑥 4 − 0.15𝑥𝑥 3 − 0.5𝑥𝑥 2 − 0.25𝑥𝑥 + 1.2
at 𝑥𝑥 = 0.5 using a step size ℎ = 0.5. Repeat the computation using ℎ = 0.25.
ENGG222: Numerical Methods in Engineering Dr. Ramin Masoudi, Mechanical Engineering 14
Finite Difference Approximations of Higher Derivatives
Second Forward Finite-divided Difference Approximation
Second Backward Finite-divided
Difference Approximation
Second Centered Finite-divided Difference Approximation
ENGG222: Numerical Methods in Engineering Dr. Ramin Masoudi, Mechanical Engineering 15
TOTAL NUMERICAL ERROR
The total numerical error is the summation of the truncation and
round-off errors
In general, the only way to minimize round-off errors is
to increase the number of significant figures of the
computer. Further, we have noted that round-off error
will increase due to subtractive cancellation or due to
an increase in the number of computations in an
analysis.
In contrast, the truncation error can be reduced by
decreasing the step size. Because a decrease in step size
can lead to subtractive cancellation or to an increase in
computations, the truncation errors are decreased as
the round-off errors are increased
ENGG222: Numerical Methods in Engineering Dr. Ramin Masoudi, Mechanical Engineering 16
