Lagrange’s Interpolation Formula

Unequally spaced interpolation requires the use of the divided difference formula. It is defined as

f (x) − f (x0 )
f (x, x0 ) = (1)
x − x0
f (x, x0 ) − f (x0 , x1 )
f (x, x0 , x1 ) = (2)
x − x1
f (x, x0 , x1 ) − f (x0 , x1 , x2 )
f (x, x0 , x1 , x2 ) = (3)
x − x2
From equation (2), the formula can be rewritten as

(x − x1 ) f (x, x0 , x1 ) + f (x0 , x1 ) = f (x, x0 ) ,

and the substitution of equation (1) yields,

(x − x0 )(x − x1 ) f (x, x0 , x1 ) + (x − x0 ) f (x0 , x1 ) + f (x0 ) = f (x) .

The first term is considered the remainder term as it is not in the difference table, so f (x) can be
expressed approximately in terms of the divided differences as

f (x) ≈ f (x0 ) + (x − x0 ) f (x0 , x1 ) + (x − x0 )(x − x1 ) f (x0 , x1 , x2 ) ,

a second order formula. The first order formula can be written as

f (x) ≈ f (x0 ) + (x − x0 ) f (x0 , x1 ) .

The above formulas are the most convenient for numerical computation when the divided differences
are store in a matrix form. But actual explicit formulas can be written in terms of the sample function
values.

Lagrange First Order Interpolation Formula

Given
f (x0 ) − f (x1 )
f (x) = f (x0 ) + (x − x0 ) .
x0 − x1
Use simplified notations f0 = f (x0 ), f1 = f (x1 ), to write

(x − x0 )
f (x) = f0 + (f1 − f0 )
(x1 − x0 )
 
(x1 − x0 ) − (x − x0 ) (x − x0 )
= f0 + f1
(x1 − x0 ) (x1 − x0 )
(x − x1 ) (x − x0 )
f (x) = f0 + f1
(x0 − x1 ) (x1 − x0 )
Lagrange Second Order Interpolation Formula

Given

f (x0 ) − f (x1 ) f (x0 , x1 ) − f (x1 , x2 )

f (x) = f (x0 ) + (x − x0 ) + (x − x0 )(x − x1 ) .
x0 − x1 x0 − x2
or  
f0 − f1 (x − x0 )(x − x1 ) f0 − f1 f1 − f2
f (x) = f0 + (x − x0 ) + −
x0 − x1 x0 − x2 x0 − x1 x1 − x2
Collecting terms for f0 , f1 and f2 , and after some tedious algebraic manipulation, the second order
formula can be written as

(x − x1 )(x − x2 ) (x − x0 )(x − x2 ) (x − x0 )(x − x1 )

f (x) = f0 + f1 + f2 .
(x0 − x1 )(x0 − x2 ) (x1 − x0 )(x1 − x2 ) (x2 − x0 )(x2 − x1 )

Lagrange N-th Order Interpolation Formula

The N-th order formula can be written in the form:

f (x) = f0 δ0 (x) + f1 δ1 (x) + . . . + fN δN (x) ,

in which, δj (x) can be written as

N
i=0;i=j (x − xi )
δj (x) = N
i=0;i=j (xj − xi )

Each term of δj (x) has the required properties such that (a) δj (xi ) = 0 when i = j and (b)
δj (xj ) = 1. The above property ensures f (xj ) = fj and none of the other sample values (fi ,
i = j) participate.