1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Lagrangian interpolation help

  1. Sep 27, 2013 #1
    1. The problem statement, all variables and given/known data
    Consider the Lagrange Polynomial approximation [tex]p(x) =\sum_{k=0}^n f(x_k)L_k(x)[/tex] where [tex]L_k(x)=\prod_{i=0,i\neq k}^n \frac{x-x_i}{x_k-x_i}[/tex]
    Let [tex]\psi(x)=\prod_{i=0}^n x-x_i[/tex]. Show that [tex]p(x)=\psi(x) \sum_{k=0}^n\frac{f(x_k)}{(x-x_k)\psi^\prime(x)}[/tex]


    2. Relevant equations
    None. Just plug in and see if it pops out.


    3. The attempt at a solution
    I just evaluated what [itex]p(x) = \sum_{k=0}^n f(x_k)L_k(x)[/itex] and [itex]\psi^\prime(x)[/itex]. Writing out some terms of p(x):[tex]p(x)=f(x_0)L_0(x_0)+f(x_1)L_1(x_1)+\cdots+f(x_n)L_n(x_n)[/tex]
    [tex]=f(x_0)\frac{\prod_{i=0,i\neq k}^n x-x_i}{\prod_{i=0,i\neq k}^n x_o-x_i}+\cdots+f(x_n)\frac{\prod_{i=0,i\neq k}^n x-x_i}{\prod_{i=0,i\neq k}^n x_n-x_i}[/tex]
    I find that this is just
    [tex]p(x)=\prod_{i=0,i\neq k}^nx-x_i \sum_{k=0}^n \frac{f(x_k)}{\prod_{i=0,i\neq k}^n x_k-x_i}[/tex]
    The product outside of the sum is just [itex]\psi(x)[/itex]. I then evaluate [itex]\psi^\prime(x)[/itex]:
    [tex]\psi^\prime(x)=\prod_{i=0,i\neq k=0}^nx-x_i +\prod_{i=0,i\neq k=1}^nx-x_i +\cdots +\prod_{i=0,i\neq k=n}^nx-x_i [/tex]
    Evaluating this at [itex]x_k[/itex] we get exactly the denominator from above. I cannot figure out where the extra [itex]x-x_k[/itex] term comes from. I suspect my error is in taking the derivative but I have looked this over for many hours and cannot find my mistake.
     
  2. jcsd
  3. Sep 27, 2013 #2
    Wow, never mind. I feel dumb. What I pulled out of the sum is not [itex]\psi(x)[/itex]. You need to multiply that product by [itex](x-x_k)[/itex] for it to be [itex]\psi(x)[/itex]. Fun with definite products - not my strong suit.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted



Similar Discussions: Lagrangian interpolation help
  1. Interpolate functions (Replies: 3)

Loading...