I'm trying to find the coefficient of $x^{2m}$ from the both sides of the following equality:

$$ \frac{(1-x^2)^n}{(1-x)^n} = (1+x)^n $$

For the right side of equality I've found it as follow:

$$ [x^{2m}] (1+x)^n=\binom{n}{2m} $$

With the left side of equation I have a problem: $$ [x^{2m}] (1-x^2)^n(1-x)^{-n}=? $$

Also, according to previous computations of coefficient, is needed to build the following equality of the sums of binomial-coefficients:

$$ \sum_{k=0}^? ??=\binom{n}{2m} $$

Can some one please explain how to get it?

For example if given that $n=5$ and $m=2$:

$$ \sum_{k=0}^n \binom{k}{k} = \binom{n}{2m}$$

$$ \sum_{k=0}^5 \binom{k}{k} = \binom{5}{4}$$

$$ \binom{0}{0} + \binom{1}{1} + \binom{2}{2} + \binom{3}{3} + \binom{4}{4} = \binom{5}{4} $$

$$ 1+1+1+1+1 = 5$$

It seems to work, but what if given $n=5$ and $m=3$? Then I'm in trouble...

What will be the correct sigma notation for this case?


  • $\begingroup$ @graydad, I think the $[x^{2m}](...)$ means the coefficient of $x^{2m}$ in $(...)$. So in this case, the OP is simply using the binomial theorem. $\endgroup$ – TravisJ Jun 15 '15 at 18:52
  • $\begingroup$ I've put $ [x^{2m}] $ just to denote that I'm looking for it's coefficient in $(a+x)^n$ $\endgroup$ – Mike Jun 15 '15 at 18:53
  • $\begingroup$ What is wrong with factoring the LHS so that it it $((1-x)(1+x))^{n}$ and cancelling. Then you know the LHS and the RHS are the same, and you know the coefficient of $x^{2m}$ in the RHS... so it must be the same on the LHS? $\endgroup$ – TravisJ Jun 15 '15 at 18:54
  • $\begingroup$ @TravisJ Yes, it must be the same as on the RHS but I need to show it, so I need to find the coefficient on the LHS as well. To my shame I have problem with factoring. and what to do with negative power of n.. Can you explain a bit how to solve the LHS? $\endgroup$ – Mike Jun 15 '15 at 19:01
  • $\begingroup$ Why not just say $\frac{(1-x^{2})^{n}}{(1-x)^{n}}=\frac{((1-x)(1+x))^{n}}{(1-x)^{n}}=(1+x)^{n} \frac{(1-x)^{n}}{(1-x)^{n}}=(1+x)^{n}$? $\endgroup$ – TravisJ Jun 15 '15 at 19:11

If you want to write $\binom{n}{m}$ as a sum of binomial coefficients, you can do it in the following way (assuming $0\leq m< n$). We will repeatedly apply Pascal's identity: $\binom{n}{m}=\binom{n-1}{m}+\binom{n-1}{m-1}$.

\begin{align*} \binom{n}{m} &= \binom{n-1}{m}+\binom{n-1}{m-1} \\ &= \binom{n-1}{m}+\binom{n-2}{m-1}+\binom{n-2}{m-2} \\ &= \binom{n-1}{m}+\binom{n-2}{m-1}+\binom{n-3}{m-2}+\binom{n-3}{m-3} \\ &\vdots \\ &=\sum_{i=0}^{m}\binom{n-(i+1)}{m-i} \end{align*}


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.