In my probability script the following equation is used, to show that for $n\to \infty, \displaystyle\binom{n}{\frac{1}{n}}$ converges against the Poisson distribution $Po(1)$. $$\lim_{n\to\infty} f_X(i) = \lim_{n\to\infty} \frac{n(n-1)...(n-i+1)}{i!}\frac{1}{n^i}\left(1-\frac{1}{n}\right)^{n-i} = \frac{1}{i!}e^{-1}$$
It is clear to me why for a set $i$, $\lim_{n\to\infty}\left(1-\frac{1}{n}\right)^{n-i} = e^{-1}$. However, I am not sure why the two first fractions result in $\frac{1}{i!}$. I thought that the $n(n-1)...(n-i+1)$ and $n!$ cancel out, but that wouldn't be an equality, at least as far as I know. It could also be that my assumption that $\lim_{n\to\infty}\left(1-\frac{1}{n}\right)^{n-i} = e^{-1}$ in this case.
\left(1\pm\frac{1}{n}\right)for the output $\left(1\pm\frac{1}{n}\right)$, so the fraction can fit in parentheses nicely and\displaystyle\binom{n}{\frac{1}{n}}for the output $\displaystyle\binom{n}{\frac{1}{n}}$. – Matcha Latte Apr 06 '20 at 07:00