Propagation of Errors

In numerical methods, the calculations are not

made with exact numbers. How do these
inaccuracies propagate through the calculations?
 Example 1:
Find the bounds for the propagation in adding two numbers. For example
if one is calculating X +Y where
X = 1.5 0.05
Y = 3.4 0.04
Solution
Maximum possible value of X = 1.55 and Y = 3.44

Maximum possible value of X + Y = 1.55 + 3.44 = 4.99

Minimum possible value of X = 1.45 and Y = 3.36.

Minimum possible value of X + Y = 1.45 + 3.36 = 4.81

Hence
4.81 ≤ X + Y ≤4.99.
 Propagation of Errors In Formulas

If f is a function of several variables X 1 , X 2 , X 3 ,......., X n −1 , X n

then the maximum possible value of the error in f is

∂f ∂f ∂f ∂f
∆f ≈ ∆X 1 + ∆X 2 + ....... + ∆X n −1 + ∆X n
∂X 1 ∂X 2 ∂X n −1 ∂X n
 Example 2:
The strain in an axial member of a square cross-
section is given by
F
∈= 2
h E
Given
F = 72 ± 0.9 N
h = 4 ± 0.1 mm
E = 70 ± 1.5 GPa

Find the maximum possible error in the measured

strain.
 Example 2:
Solution
72
∈= −3 2
(4 × 10 ) (70 × 10 )
9

= 64.286 × 10 −6
= 64.286 µ

∂∈ ∂∈ ∂∈
∆ ∈= ∆F + ∆h + ∆E
∂F ∂h ∂E
 Example 2:
∂∈ 1 ∂∈ 2F ∂∈ F
= 2 =− 3 =− 2 2
∂F h E ∂h hE ∂E h E
Thus
1 2F F
∆E = 2 ∆F + 3 ∆h + 2 2 ∆E
h E hE h E
1 2 × 72
= −3 2
× 0 .9 + −3 3
× 0.0001
(4 ×10 ) (70 ×10 )
9
(4 ×10 ) (70 ×10 )
9

72
+ × 1 .5 × 10 9

(4 ×10 −3 ) 2 (70 ×109 ) 2

= 5.3955µ
Hence
∈= (64.286 µ ± 5.3955µ )
 Example 3:
Subtraction of numbers that are nearly equal can create unwanted
inaccuracies. Using the formula for error propagation, show that this is true.

Solution
Let
z = x− y
Then
∂z ∂z
∆z = ∆x + ∆y
∂x ∂y
= (1)∆x + (−1)∆y
= ∆x + ∆y
So the relative change is
∆z ∆x + ∆y
=
z x− y
 Example 3:
For example if
x = 2 ± 0.001
y = 2.003 ± 0.001

∆z 0.001 + 0.001
=
z | 2 − 2.003 |
= 0.6667
= 66.67%
 Taylor Theorem
Motivation
 We can easily compute expressions like:
3 10 2
2( x  4)
But, How do you compute 4.1, sin(0.6) ?

Can we use the definition

b
a
to compute sin(0.6)?
0.6
Is this a practical way?
Taylor Series
The Taylor series expansion of f ( x ) about a :
(2) ( 3)
f ( a ) f (a )
f (a )  f ' (a ) ( x  a )  ( x  a )2  ( x  a ) 3  ...
2! 3!
or

1 (k )
Taylor Series   k!
f (a ) ( x  a )k
k 0
If the series converge, we can write :
∞
1 (k )
f ( x)  ∑ f (a ) ( x  a )k
k 0
k!
Maclaurin Series
 Maclaurin series is a special case of Taylor
series with the center of expansion a = 0.
The Maclaurin series expansion of f ( x ) :
( 2) ( 3)
' f ( 0 ) 2 f ( 0) 3
f ( 0 )  f ( 0) x  x  x  ...
2! 3!
If the series converge, we can write :
∞
1 (k ) k
f ( x)  ∑ f ( 0) x
k 0
k!
Maclaurin Series – Example 1
Obtain Maclaurin series expansion of f ( x )  e x
f ( x)  e x f ( 0)  1
f ' ( x)  e x f ' ( 0)  1
f ( 2) ( x )  e x f ( 2 ) ( 0)  1
f (k ) ( x)  e x f ( k ) (0)  1 for k  1
∞ ∞ k 2 3
1 x x x
ex  ∑ f ( k ) ( 0) x k  ∑  1 x    ...
k 0
k! k 0
k! 2! 3!
The series converges for x  ∞.
Taylor Series
3

Example 1
2.5
exp(x)
1+x+0.5x 2
2

1+x

1.5

1
1

0.5

0
-1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1
Maclaurin Series – Example 2
Obtain Maclaurin series expansion of f ( x )  sin( x ) :
f ( x )  sin( x ) f ( 0)  0
f ' ( x )  cos( x ) f ' ( 0)  1
( 2) ( 2)
f ( x )   sin( x ) f ( 0)  0
f ( 3) ( x )   cos( x ) f ( 3) (0)  1
∞
f ( k ) ( 0) k x3 x5 x7
sin( x )  ∑ x  x     ....
k 0
k! 3! 5! 7!
The series converges for x  ∞.
4

3
x

1 x-x 3/3!+x 5/5!

0 sin(x)

-1
x-x 3/3!

-2

-3

-4
-4 -3 -2 -1 0 1 2 3 4
Maclaurin Series – Example 3
Obtain Maclaurin series expansion of : f ( x)  cos( x)
f ( x )  cos( x ) f (0)  1
f ' ( x )   sin( x ) f ' ( 0)  0
f ( 2 ) ( x )   cos( x ) f ( 2 ) ( 0)  1
f ( 3) ( x )  sin( x ) f ( 3) (0)  0
∞
f ( k ) ( 0) x 2
x 4
x 6
cos( x )  ∑ ( x ) k  1     ....
k 0
k! 2! 4! 6!
The series converges for x  ∞.
Maclaurin Series – Example 4
1
Obtain Maclaurin series expansion of f(x) 
1 x
1
f ( x)  f ( 0)  1
1 x
1
f ' ( x)  f ' ( 0)  1
1  x 2

(2) 2 ( 2)
f ( x)  f ( 0)  2
1  x 3

( 3) 6 ( 3)
f ( x)  f ( 0)  6
1  x  4

1
Maclaurin Series Expansion of :  1  x  x 2  x 3  ...
1 x
Series converges for | x |  1
Example 4 - Remarks

 Can we apply the series for x≥1??

 How many terms are needed to get a good

approximation???

These questions will be answered using

Taylor’s Theorem.
Taylor Series – Example 5
1
Obtain Taylor series expansion of f(x)  at a  1
x
1
f ( x)  f (1)  1
x
1
f ' ( x)  2 f ' (1)  1
x
2
f ( 2) ( x )  3 f ( 2 ) (1)  2
x
( 3) 6
f ( x)  4 f ( 3) (1)  6
x
Taylor Series Expansion (a  1) : 1  ( x  1)  ( x  1) 2  ( x  1)3  ...
Taylor Series – Example 6
Obtain Taylor series expansion of f(x)  ln( x ) at ( a  1)

1 ( 2) 1 ( 3) 2
f ( x )  ln( x ) , f ' ( x )  , f ( x )  2 , f ( x )  3
x x x
f (1)  0, f ' (1)  1, f ( 2 ) (1)  1 f ( 3) (1)  2

1 2 1
Taylor Series Expansion : ( x  1)  ( x  1)  ( x  1) 3  ...
2 3
Convergence of Taylor Series

 The Taylor series converges fast (few terms

are needed) when x is near the point of
expansion. If |x-a| is large then more terms
are needed to get a good approximation.
Taylor’s Theorem
If a function f ( x ) possesses derivatives of orders 1, 2, ..., ( n  1)
on an interval containing a and x then the value of f ( x ) is given by :

(n+1) terms Truncated

Taylor Series
n
f ( k ) (a )
f ( x)  ∑ ( x  a)k  Rn
k 0
k!
Remainder
where :
f ( n 1) ( )
Rn  ( x  a ) n 1 and  is between a and x.
( n  1)!
Taylor’s Theorem
We can apply Taylor' s theorem for :
1
f(x)  with the point of expansion a  0 if | x |  1.
1 x

If x  1, then the function and its

derivatives are not defined.
 Taylor Theorem is not applicable.
Error Term

To get an idea about the approximation error,

we can derive an upper bound on :
( n 1)
f ( )
Rn  ( x  a ) n 1
( n  1)!
for all values of  between a and x.
Error Term - Example
x
How large is the error if we replaced f ( x )  e by
the first 4 terms ( n  3) of its Taylor series expansion
at a  0 when x  0.2 ?
(n) x (n) 0.2
f ( x)  e f ( ) ≤ e for n ≥ 1
( n 1)
( )
f n 1
Rn  ( x  a)
( n  1)!
e 0.2
Rn  0.2 n 1  R3  8.14268E  05
( n  1)!
Alternative form of Taylor’s Theorem
Let f ( x ) have derivatives of orders 1, 2, ..., ( n  1)
on an interval containing x and x  h then :
n
f (k ) ( x) k
f ( x  h)   k!
h  Rn ( h  step size)
k 0

( n 1)
f ( ) n 1
Rn  h where  is between x and x  h
( n  1)!
Taylor’s Theorem – Alternative forms
n ( n 1)
f ( k ) (a ) f ( )
f ( x)   k
( x  a)  ( x  a ) n 1
k 0 k! ( n  1)!
where  is between a and x.

a  x, x  x  h

n
f ( k ) ( x ) k f ( n 1) ( ) n 1
f ( x  h)   h  h
k 0 k! ( n  1)!
where  is between x and x  h.
Mean Value Theorem
If f ( x ) is a continuous function on a closed interval [a , b]
and its derivative is defined on the open interval ( a , b)
then there exists ξ  ( a, b)
f(b)  f(a)
f ' (ξ ) 
ba
Proof : Use Taylor' s Theorem for n  0, x  a , x  h  b
f(b)  f(a)  f ' (ξ ) (b  a )
Alternating Series Theorem
Consider the alternating series :
S  a1  a2  a3  a4  
 a  a  a  a  The series converges
 1 2 3 4

If  and then  and
 lim a  0  S  S n  an 1
 n n 

S n : Partial sum (sum of the first n terms)

an 1 : First omitted term
Alternating Series – Example
1 1 1
sin(1) can be computed using : sin(1)  1     
3! 5! 7!
This is a convergent alternating series since :
a1  a2  a3  a4   and lim an  0
n 

Then :
 1 1
sin(1)  1   
 3!  5!
 1 1 1
sin(1)  1    
 3! 5!  7!
Example 7

Obtain the Taylor series expansion

2 x 1
of f ( x )  e at a  0.5 (the center of expansion)
How large can the error be when ( n  1) terms are used
to approximate e 2 x 1 with x  1 ?
Example 7 – Taylor Series
Obtain Taylor series expansion of f ( x )  e 2 x 1 , a  0.5
f ( x)  e 2 x 1 f (0.5)  e 2
f ' ( x)  2e 2 x 1 f ' (0.5)  2e 2
f ( 2) ( x)  4e 2 x 1 f ( 2) (0.5)  4e 2
f ( k ) ( x)  2 k e 2 x 1 f ( k ) (0.5)  2 k e 2
f ( k ) (0.5)
∞

e 2 x1 ∑ ( x  0.5) k
k 0
k!
2 k
( x  0 .5) ( x  0 . 5)
 e 2  2e 2 ( x  0.5)  4e 2  ...  2 k e 2  ...
2! k!
Example 7 – Error Term
f ( k ) ( x)  2 k e 2 x 1
f ( n 1) ( )
Error  ( x  0.5) n 1
(n  1)!
(1  0.5) n 1
Error  2 n 1 e 2 1
(n  1)!
n 1
( 0. 5)
Error  2 n 1 max e 2 1
(n  1)!  [ 0.5,1]
e3
Error 
(n  1)!