Re: HSC 2013 4U Marathon
![](https://latex.codecogs.com/png.latex?\bg_white $i)$ \sum_{r=0}^{n}(r+1)^k - r^k = (n+1)^k)
by a telescoping sum.
Now expanding the binomial first instead of using a telescoping sum gives:
Now if we apply the summation, we sum each one individually, that is:
Now using our definition of
![](https://latex.codecogs.com/png.latex?\bg_white S_k(n))
we obtain:
Using the symmetry property of binomial expansions gives:
Now since
![](https://latex.codecogs.com/png.latex?\bg_white \binom{k}{1} = k)
and
![](https://latex.codecogs.com/png.latex?\bg_white \binom{k}{0} = 1)
we get:
Now consider the second part of the equation:
![](https://latex.codecogs.com/png.latex?\bg_white \sum_{r=0}^{n} -r^k = -\sum_{r=0}^{n} r^k = -S_k(n))
by definition.
Hence if we add the two parts together we obtain:
So the two terms on the end cancel out and give:
But this was equal to the value of the telescoping sum, so we can equate the two and get:
Taking everything from the RHS to the LHS except for the last term gives us the required result of:
Which can be re-written in the form asked:
Quod Erat Demonstrandum.