1
$\begingroup$

I have a formula where both summation and product are involved \begin{align*} f(n-1)=\sum_{m=0}^{n-1}\sum_{j=0}^{m}\frac{\binom{m}{j}(-1)^j\mathrm{i}^m (a+k)^{2m-n+1}}{m!(2 a)^m} \prod_{p=1}^{n-1}(2m-2j-p+1). \end{align*} I want that the summation and product runs until $ n $ instead $ n-1 $. i.e., $ f(n)$. What changes will be made? I have no idea, how to do this.

  • 0
    It seems you just want an expression for $f(n)$. If so, just plug in $n$ instead of $n-1$... Would you like me to write up what I mean? – 2017-01-25
  • 0
    yes....If possible. My main aim is to run the formula for $ n=0$ as well. In current situation, it can run from $ n=1$ – 2017-01-25
  • 0
    I will write up a solution. One moment. – 2017-01-25
  • 0
    There. Helpful? – 2017-01-25
  • 0
    Yes....Thanks a lot :) – 2017-01-25
  • 0
    Actually my problem is still there....This is part of $ n $ th derivative of error function. When I put $ n=0$, it should give original function, unfortunately its not. – 2017-01-25
  • 0
    Well, what I wrote is in fact the answer to the question you asked. – 2017-01-25
  • 0
    Yes....can you please help me in that question? I tried to change limits but it is not working. – 2017-01-25
  • 0
    But now I don't know what your question is. You asked for $f(n)$ and that is what I solved for. – 2017-01-25
  • 0
    How to send you my question? should I post again and then send a link to you? – 2017-01-25
  • 0
    ok.....Let me edit this question and write the exact problem. – 2017-01-25
  • 0
    @TheCount this is my exact problem http://math.stackexchange.com/questions/2114167/leibniz-rule-for-error-function – 2017-01-25

1 Answers 1

0

You already have this formula for $f(n-1)$:

$$f(n-1)=\sum_{m=0}^{n-1}\sum_{j=0}^{m}\frac{\binom{m}{j}(-1)^j\mathrm{i}^m (a+k)^{2m-n+1}}{m!(2 a)^m} \prod_{p=1}^{n-1}(2m-2j-p+1).$$

And since you want $f(n)$, we just need to replace every instance of $n-1$ with $n$. But this is a little tricky. There are three things to tackle. The first two are easy: the upper limits on the sum and product become $n$ instead of $n-1$. But on your factor of $(a+k)^{2m-n+1}$, we also have an $n-1$ hiding. The exponent needs to become $2m-(n+1)+1=2m-n$. Do you see why? In any case, our final expression is:

$$f(n)=\sum_{m=0}^{n}\sum_{j=0}^{m}\frac{\binom{m}{j}(-1)^j\mathrm{i}^m (a+k)^{2m-n}}{m!(2 a)^m} \prod_{p=1}^{n}(2m-2j-p+1).$$