ME 261

Session: January 2023

Dr. Sumon Saha

Professor
Department of Mechanical Engineering
Bangladesh University of Engineering and Technology
ME 261 .. Numerical Analysis

Lecture # 07

Numerical Interpolation
 ME 261 .. Introduction to Interpolation

• Interpolation Function: A function that passes exactly

through a set of data points.
• Interpolating functions to interpolate values in Tables

• In Table, the function is only specified at a limited number

or discrete set of independent variable values (as opposed to
a continuum function)
• We can use interpolation to find functional values at other
values of the independent variable, e.g., sin(0.63253)
 ME 261 .. Introduction to Interpolation

• In numerical methods, like Tables, the values of the

function are only specified at a discrete number of points!
Using interpolation, we can describe or at least approximate
the function at every point in space.
• For numerical methods, we use interpolation techniques for
the following cases:
– Interpolate values from computations
– Develop numerical integration schemes
– Develop numerical differentiation schemes
– Develop finite element methods
• Interpolation is typically not used to obtain a functional
description of measured data since errors in the data may
lead to a poor representation.
– Curve fittings to data is handled with a separate set of techniques
ME 261 .. Regression versus Interpolation
 ME 261 .. Polynomial interpolation

• Estimation of intermediate values between precise data

points. The most common method is:
n −1
f ( x) = a1 + a2 x + a3 x + 2
+ an x
• Although there is one and only one (n-1)th-order polynomial
that fits n points, there are a variety of mathematical formats
in which this polynomial can be expressed:
– The Newton polynomial
– The Lagrange polynomial
• Although both Newton and Lagrange polynomials are well
suited for determining intermediate values between points,
they do not provide a polynomial in the above conventional
form.
 ME 261 .. Coefficients of an Interpolating Polynomial

Since n data points are required to determine n coefficients,

simultaneous systems of linear equations can be used to
calculate “a”s.

f ( x1 ) = a1 + a2 x1 + a3 x12 + an x1n −1
n −1
f ( x2 ) = a1 + a2 x2 + a x 2
3 2 +a x n 2

n −1
f ( xn ) = a1 + a2 xn + a x 2
3 n +a x n n

Where “x”s are the knowns,

and “a”s are the unknowns.
ME 261 .. Coefficients of an Interpolating Polynomial
 ME 261 .. Polynomial interpolation

Examples of interpolating polynomials: (a) first-order (linear)

connecting two points, (b) second-order (quadratic or parabolic)
connecting three points, and (c) third-order (cubic) connecting
four points.
 ME 261 .. Newton’s Divided-Difference Interpolating
Polynomials
Linear Interpolation/
• Is the simplest form of interpolation, connecting two data
points with a straight line.
Slope and a
f1 ( x) − f ( x1 ) f ( x2 ) − f ( x1 ) finite divided
= difference
x − x1 x2 − x1 approximation to
f ( x2 ) − f ( x1 ) 1st derivative
f1 ( x) = f ( x1 ) + ( x − x1 )
x2 − x1 Linear-interpolation
formula

• f1(x) designates that this is a first-order interpolating

polynomial.
 ME 261 .. Linear Interpolation

Taking slopes for both

triangles
f1 ( x) − f ( x1 ) f ( x2 ) − f ( x1 )
=
x − x1 x2 − x1

f ( x2 ) − f ( x1 )
f1 ( x) = f ( x1 ) + ( x − x1 )
x2 − x1
 ME 261 .. Quadratic Interpolation

• If three data points are available, the estimate is improved

by introducing some curvature into the line connecting the
points.
f 2 ( x) = b1 + b2 ( x − x1 ) + b3 ( x − x1 )( x − x2 )
• A simple procedure can be used to determine the values of
the coefficients.
x = x1 b1 = f ( x1 )
f ( x2 ) − f ( x1 )
x = x2 b2 =
x − x1
f ( x3 ) − f ( x2 ) f ( x2 ) − f ( x1 )
−
x3 − x2 x2 − x1
x = x3 b3 =
x3 − x1
 ME 261 .. General Form of Newton’s Interpolating
Polynomials
f n −1 ( x) = b1 + b2 ( x − x1 ) + b3 ( x − x1 )( x − x2 ) + + bn ( x − x1 )( x − x2 ) ( x − xn −1 )
f n −1 ( x) = f ( x1 ) + ( x − x1 ) f [ x2 , x1 ] + ( x − x1 )( x − x2 ) f [ x3 , x2 , x1 ] +
+( x − x1 )( x − x2 ) ( x − xn −1 ) f [ xn , xn −1 , , x1 ]
b1 = f ( x1 )
b2 = f [ x2 , x1 ]
b3 = f [ x3 , x2 , x1 ]

bn = f [ xn , xn −1 , , x 2 , x1 ]
f ( xi ) − f ( x j )
f [ xi , x j ] =
xi − x j
f [ xi , x j ] − f [ x j , xk ] Bracketed function evaluations are
f [ xi , x j , xk ] = finite divided differences
xi − xk

f [ xn , xn −1 , , x2 ] − f [ xn −1 , xn − 2 , , x1 ]
f [ xn , xn −1 , , x2 , x1 ] =
xn − x1
 ME 261 .. General Form of Newton’s Interpolating
Polynomials

f ( xi ) − f ( x j ) f [ xi , x j ] − f [ x j , xk ]
f [ xi , x j ] = f [ xi , x j , xk ] =
xi − x j xi − xk
Question: The general form of Newton’s interpolating polynomial is-
f n ( x) = b0 + b1 ( x − x0 ) + b2 ( x − x0 )( x − x1 ) + + bn ( x − x0 )( x − x1 ) ( x − xn −1 )
Show that the coefficients b0, b1, b2,…, bn actually indicate derivates by
Newton’s Divided Difference Method.
 ME 261 .. Linear Interpolation

Linear interpolation is obtained by passing a straight line between

2 data points
x1, x2 = the data points (also
f(x) referred to as interpolation points
or nodes)
f1(x) f(x) = exact function for which
values are known only at a
discrete set of data points

f1(x) = the interpolated

approximation to f(x)
 ME 261 .. Linear Interpolation

If f1(x) is a linear function with two unknown coefficients A and B

f1 ( x) = Ax + B

To pass through points (x1, f(x1)) and (x2, f(x2)) we must have:
f1 ( x1 ) = f ( x1 )  Ax1 + B = f ( x1 )
f1 ( x2 ) = f ( x2 )  Ax2 + B = f ( x2 )
2 unknowns and 2 equations  Solve for A and B
B = f ( x1 ) − Ax1
Ax2 + f ( x1 ) − Ax2 = f ( x2 )
B

f ( x2 ) − f ( x1 )
A=
x2 − x1
f ( x1 ) x2 − f ( x2 ) x1
B=
x2 − x1
 ME 261 .. Linear Interpolation

Now substituting the value for A and B in the approximating

polynomial
( x2 − x) ( x − x1 ) ( x − x2 ) ( x − x1 )
f1 ( x) = f ( x1 ) + f ( x2 ) = f ( x1 ) + f ( x2 )
( x2 − x1 ) ( x2 − x1 ) ( x1 − x2 ) ( x2 − x1 )

This is the formula for linear interpolation.

f ( x2 ) − f ( x1 )
f1 ( x) = f ( x1 ) + ( x − x1 )
x2 − x1
This is Newton’s Divided-Difference first-order interpolating
polynomial.

Question: Derive Newton’s first order interpolating polynomial. Also

derive the Lagrange form from that interpolating polynomial.
 ME 261 .. Lagrange Interpolating Polynomial

Linear interpolating polynomial  weighted average of the two

values that are connected by a straight line
f ( x) = L1 f ( x1 ) + L2 f ( x2 )

Where, L’s are the weighing coefficients defined such that it returns
1 at one point and 0 at the other point
x − x2
L1 = 1 at x1 and 0 at x2
x1 − x2
x − x1
L2 = 0 at x1 and 1 at x2
x2 − x1
Substituting these into the interpolating
polynomial x − x2 x − x1
f ( x) = f ( x1 ) + f ( x2 )
x1 − x2 x2 − x1
Linear Lagrange interpolating polynomial
 ME 261 .. Lagrange Interpolating Polynomial

Linear Lagrange interpolating polynomial (for first order)

x − x2 x − x1
f1 ( x) = f ( x1 ) + f ( x2 )
x1 − x2 x2 − x1

Second order Lagrange interpolating polynomial => fit a parabola

through three points (x , x , x ) 1 2 3

Weighing coefficients are parabolas giving one passing through one

point and zero for other two.
Now, the weighted average of the three values that are connected by a
unique parabola through the three points:
( x − x2 )( x − x3 ) ( x − x1 )( x − x3 ) ( x − x1 )( x − x2 )
f 2 ( x) = f ( x1 ) + f ( x2 ) + f ( x3 )
( x1 − x2 )( x1 − x3 ) ( x2 − x1 )( x2 − x3 ) ( x3 − x1 )( x3 − x2 )

Second order Lagrange interpolating polynomial

 ME 261 .. Lagrange Interpolating Polynomial

For higher order Lagrange polynomial

n
f n−1 ( x) =  Li ( x) f ( xi )
i =1

Where n x − xj
Li ( x) = 
j =1 xi − x j
j i

where n is the number of data points and  designates “the product of”

• The Lagrange interpolating polynomial is simply a

reformulation of the Newton’s polynomial that avoids
the computation of divided differences.
 ME 261 .. Numerical Interpolation

Example 01:
Estimate the natural logarithm of 2 using linear interpolation.
First perform the computation by interpolating between ln 1 =
0 and ln 6 = 1.791759. Then, repeat the procedure, but use a
smaller interval from ln 1 to ln 4 (1.386294). Note that the true
value of ln 2 is 0.6931472.
Solution: For linear interpolation,
f ( x2 ) − f ( x1 )
f1 ( x) = f ( x1 ) + ( x − x1 )
x2 − x1
 ME 261 .. Numerical Interpolation

Solution: From x1 = 1 to x2 = 6,
f (6) − f (1) 1.791759 − 0
f1 (2) = f (1) + (2 − 1) = 0 + ( 2 − 1) = 0.3583519
6 −1 6 −1
fexact ( 2 ) − f num ( 2 ) 0.6931472 − 0.3583519
Error, t = 100% = 100% = 48.3%
fexact ( 2 ) 0.6931472

From x1 = 1 to x2 = 4,
f (4) − f (1) 1.386294 − 0
f1 (2) = f (1) + (2 − 1) = 0 + ( 2 − 1) = 0.4620981
4 −1 4 −1
Error, εt = 33.3%
 ME 261 .. Numerical Interpolation

Two linear interpolations to estimate ln 2. Note how the smaller interval

provides a better estimate.
 ME 261 .. Numerical Interpolation

Example 02:
Employ a second-order Newton polynomial to estimate ln 2
with the following three points. Note that the true value of ln 2
is 0.6931472.
x 1 4 6
f(x) 0 1.386294 1.791759
 ME 261 .. Numerical Interpolation

Solution: Divided Difference Table

x f(x) First Second
1 0 0.4620981 −0.0518731
4 1.386294 0.2027325
6 1.791759

f2 ( 2) = 0 + 0.4620981( 2 − 1) − 0.0518731( 2 − 1)( 2 − 4 )

= 0.5658444

Error, εt = 18.4%
 ME 261 .. Numerical Interpolation

Two linear interpolations to estimate ln 2. The linear interpolation from x

= 1 to 4 is also included for comparison.
 ME 261 .. Numerical Interpolation

Example 03:
Employ a third-order Newton’s interpolating polynomial to
estimate ln 2 with the following four points. Note that the true
value of ln 2 is 0.6931472.
x 1 4 5 6
f(x) 0 1.386294 1.609438 1.791759
 ME 261 .. Numerical Interpolation

Solution: Divided Difference Table

x f(x) First Second Third
1 0 0.4620981 −0.0597385 0.0078654
4 1.386294 0.223144 −0.0204115
5 1.609438 0.182321
6 1.791759

f3 ( 2 ) = 0 + 0.4620981( 2 − 1) − 0.0597385 ( 2 − 1)( 2 − 4 ) + 0.0078654 ( 2 − 1)( 2 − 4 )( 2 − 5 )

= 0.6287675

Error, εt = 9.3%
ME 261 .. Numerical Interpolation

Two linear interpolations to estimate ln 2.