Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Sum of a series

  1. Sep 20, 2006 #1
    Let be the series....[tex] S=\sum_{n=0}^{\infty} a(n) [/tex]

    where a(0)=1=a(1) and the rest of coefficients satisfy a recurrence relation (linear or non-linear) so [tex] F(n,a_{n+2} , a_{n+1},a_{n})=n [/tex] :tongue2: :tongue2: ..then my question is let's suppose that the series has an "optimum number of terms" K so if you take k-terms the series converges to a optimum value, otherwise the series (taking all terms) diverges) my question is how would we obtain this k and the sum of the series... a "brute force" algorithm would say that you take a big number of terms and solve the recurrence by using a computer...:rolleyes: :rolleyes:
  2. jcsd
  3. Sep 20, 2006 #2

    matt grime

    User Avatar
    Science Advisor
    Homework Helper

    Since you have not defined optimum the question is impossible to answer.

    When you use K and k are we supposed to think they refer to the same thing? What is a k-term? Do you just mean sum the first k terms (so why use the word convergent for a finite sum?), or do you mean to pick some infinite subset of the terms?

    Your recurrence relation could very well be easy to solve (it is only a second order recurrence relation, as written.
    Last edited: Sep 20, 2006
  4. Sep 20, 2006 #3


    User Avatar
    Science Advisor
    Homework Helper

    The sum of a finite number of finite terms always converges. If your ininite sum is divergent, then any stopping point will put you in the situation you mention.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook