explicit formula for divided differences


Theorem 1.

The n-th divided differenceDlmfMathworldPlanetmath of a functionMathworldPlanetmath f can be written explicitly as

Δnf[x0,,xn]=i=0nf(xi)0jnji(xi-xj)
Proof.

We will proceed by recursion on n. When n=1, the formula to be proven reduces to

Δ1f[x0,x1]=f(x0)x0-x1+f(x1)x1-x0,

which agrees with the definition of Δ1f.

To prove that this is correct when n>1, one needs to check that it the recurrence relation for divided differences.

0in+1i0f(xi)0jn+1jij0(xi-xj) -0in+1i1f(xi)0jn+1jij1(xi-xj)
=f(x1)0jn+1jij0(x1-xj)-f(x0)0jn+1jij1(x0-xj)+
  2in+1f(xi)(10jn+1jij0(xi-xj)-10jn+1jij1(xi-xj))
=-(x0-x1)f(x0)0jn+1ji(x0-xj)+(x1-x0)f(x1)0jn+1ji(x1-xj)+
  2in+1f(xi)(xi-x00jn+1ji(xi-xj)-xi-x10jn+1ji(xi-xj))
=(x1-x0)f(x0)0jn+1ji(x0-xj)+(x1-x0)f(x1)0jn+1ji(x1-xj)+(x1-x0)2in+1(x1-x0)f(xi)0jn+1ji(xi-xj)
=(x1-x0)0in+1f(xi)0jn+1ji(xi-xj)

Thus, we see that, if

Δnf[x0,,xn]=i=0nf(xi)0jnji(xi-xj),

then

Δn+1f[x0,,xn+1]=i=0n+1f(xi)0jn+1ji(xi-xj).

Hence, by induction, the formula holds for all n. ∎

This formula may be phrased another way by introducing the polynomials pn defined as

pn(x)=i=0n(x-xi).

We may write

Δnf[x0,,xn]=i=0nf(xi)p(xi).

Either form of the explicit formula makes it obvious that divided differences are symmetric functions of x0,x1,.

Title explicit formula for divided differences
Canonical name ExplicitFormulaForDividedDifferences
Date of creation 2013-03-22 14:41:16
Last modified on 2013-03-22 14:41:16
Owner rspuzio (6075)
Last modified by rspuzio (6075)
Numerical id 21
Author rspuzio (6075)
Entry type Definition
Classification msc 39A70