Proving (a+b)^n = ∑(n_μ)(a^μ)(b^(n-μ)) | a=1, b=-1 using Sequences Test

  • Thread starter Thread starter AnthonyAcc
  • Start date Start date
  • Tags Tags
    Sequences Test
AnthonyAcc
Messages
5
Reaction score
0
Show that (^{n}_{n}) - (^{n}_{n-1}) + (^{n}_{n-2}) - (^{n}_{n-3}) + ...(^{n}_{0}) = 0

(a+b)^{n} = \sum^{\infty}_{\nu=0} (^{n}_{\nu})a^{\nu}b^{n-\nu}a=1
b=-1

0 = (1+(-1))^{n} = \sum^{\infty}_{\nu=0}(^{n}_{\nu}) 1^{\nu}(-1)^{n-\nu} = \sum^{\infty}_{\nu=0}(^{n}_{\nu})(-1)^{n-\nu} =

...I don't know what to do here...

= \sum^{n}_{\nu=0}(^{n}_{n-\nu})(-1)^{\nu} = (^{n}_{n}) - (^{n}_{n-1}) + (^{n}_{n-2}) - (^{n}_{n-3}) + ...(^{n}_{0})That is if that last equality is correct and makes sense to be there.

Any suggestions?

Also, how can I make it so everything that follows the Sigma doesn't look like it is being superscripted?

Thanks!
 
Last edited:
Physics news on Phys.org
a=1, b=-1
(1+(-1))n=0

Now try a=-1, b=1 so that

(-1+1)n=0

Or separate the odds from the even and create two separate sums:
sum (even) - sum (odd) = 0
 
Does this \sum^{\infty}_{\nu=0} (^{n}_{\nu}) imply \sum^{n}_{\nu=0} (^{n}_{\nu}) because it's for every n choose \nu so there can only be n many \nus?
 
I told you to use something like this:
\sum_{v=0}^{n/2}\binom{n}{n-2v}(-1)^{2v}*1^{n-2v}+\sum_{v=0}^{n/2}\binom{n}{n-2v-1}(-1)^{2v-1}*1^{n-2v-1}

For the both sums a=-1 and b=1, so that 0-0=0

Use \binom{ } { } for binomial coefficients.

Regards.
 
Last edited:
A more fundamental question I have, I guess, is if I know that \sum^{\infty}_{\nu=0} (^{n}_{\nu}) a^{\nu} b^{n-\nu} = (a+b)^{n} then do I know that
\sum^{n}_{\nu=0} (^{n}_{n-\nu}) a^{\nu} b^{n-\nu} = (a+b)^{n}? How does the changing of infinity to n and the \nu to n - \nu impact the series?
 
Last edited:
n is an integer, right? Then the generalized binomial coefficient C(n,nu) of n and nu is zero for nu>n, also right? There's a zero factor in the definition. That's how you truncate the infinite sum. Though I'm not sure why you have an infinite sum to begin with. And in that case, sure, C(n,nu)=C(n,n-nu).
 
Back
Top