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

Factorial question in a power series solution

  1. Oct 30, 2011 #1
    Hello, I've been working on solving the equation y''-2xy'+2py=0. where p is a positive integer.

    I've assumed y=[itex]\sum a_{n}x^{n}[/itex] for n=0 to inf

    I'm getting two formulas for [itex]a_{n}[/itex]
    One is for odd n, the other for even n, related to [itex]a_{0}[/itex] and [itex]a_{1}[/itex]

    However, the relation involves something that looks like a factorial but it skips every other number, such as p(p-2)(p-4)(p-6)...

    My question is whether this is valid:

    [itex]n!=n(n-1)(n-2)(n-3)...[/itex]

    [itex]2^{n}(n)!=2n(2n-2)(2n-4)(2n-6)...[/itex]

    letting [itex]p=2n [/itex] then

    [itex]2^{p-1}(\frac {p}{2})!=p(p-2)(p-4)(p-6)...[/itex]
     
  2. jcsd
  3. Oct 30, 2011 #2

    HallsofIvy

    User Avatar
    Staff Emeritus
    Science Advisor

    Yes, that is perfectly valid. I would have analyzed it the other way:
    [tex]2*4*6*...*(2n)= (2*1)(2*2)(2*3)...(2*n)= 2^n(n!)[/tex]
    You can also say that
    [tex]1*3*5*...*(2n+1)= \frac{1*2*3*4*5*...*(2n)*(2n+1)}{2*4*...*(2n)}= \frac{(2n+1)!}{2^n n!}[/tex]
     
  4. Oct 30, 2011 #3
    I am supposed to find the solution in summation form, would you mind looking over it? I'm not sure if the factorials give the correct terms based on n. I'll post my equations for the coefficients, and what I think is the formula for them.

    n even:
    [itex]a_{2}=-pa_{0}[/itex]

    [itex]a_{4}=\frac{(-1)^{2}2^{2}p(p-2)a_{0}}{4!}[/itex]

    [itex]a_{6}=\frac{(-1)^{3}2^{3}p(p-2)(p-4)a_{0}}{6!}[/itex]

    [itex]a_{8}=\frac{(-1)^{4}2^{4}p(p-2)(p-4)(p-6)a_{0}}{8!}[/itex]

    n odd:
    [itex]a_{3}=\frac{(-1)2(p-1)a_{1}}{3!}[/itex]

    [itex]a_{5}=\frac{(-1)^{2}2^{2}(p-1)(p-3)a_{1}}{5!}[/itex]

    [itex]a_{7}=\frac{(-1)^{3}2^{3}(p-1)(p-3)(p-5)a_{1}}{7!}[/itex]


    my guess at the 'summation form of the complete solution' is this:

    [itex]y(x)=-pa_{0} + \sum a_{n}x^{n}[/itex] for n = 1 to inf

    where [itex]a_{n} = \frac {-1^{\frac{n}{2}}2^{\frac{n}{2}}(\frac{p}{2})!a_{0}}{n!(\frac{p-n}{2})!}[/itex] for n even

    and [itex]a_{n} = \frac {-1^{\frac{n-1}{2}}2^{\frac{n-1}{2}}(\frac{p-1}{2})!a_{1}}{n!(\frac{p-n-1}{2})!}[/itex] for n odd

    I know that's a nightmare, but I'd appreciate it if you could criticize or tell me if this is wrong / if its acceptable in form. I don't know that the ratio of the factorials gives the terms I need above, I think it does.
     
    Last edited: Oct 30, 2011
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Factorial question in a power series solution
Loading...