# Evaluation of $\lim _{n\rightarrow \infty }\left( e^{-na}n^{b}\right)$ when $a > 0, b > 0$

Does $$\lim _{n\rightarrow \infty }\left( e^{-na}n^{b}\right)$$ evaluate to $\infty$ when $a > 0, b > 0$.

I tried the expansion of $e^{-na}$ but could not shake of n from numerators.

-
Think of the expression as $n^b\over e^{na}$ and note all powers are positive. Which wins out, exponentials or powers? – David Mitra Mar 6 '12 at 19:17
it should evaluates to 0, since for a large number exponential is way larger than polynomial. – quartz Mar 6 '12 at 19:19

To try and complete your attempt:

You can use the expansion of $e^n$ to show that for any $c \gt 0$, $e^n \gt Kn^c$ for some constant $K \gt 0$ (dependent on $c$).

Let $[c] = m-1$ ($[x]$ is the integer part of $x$).

Since $e^n = 1+ n + \frac{n^2}{2} + \dots + \frac{n^{m}}{m!} + \dots \gt \frac{n^m}{m!}$

Now $n^m \gt n^c$ and so $e^n \ge \frac{n^c}{m!}$

In your case, we can pick $c = \frac{b+1}{a}$.

So we get

$$e^n \ge K n^{\frac{b+1}{a}}$$

i.e

$$e^{na} \ge K^a n^{b+1}$$

and

so

$$\frac{n^b}{e^{na}} \le \frac{1}{K^a n}$$

And so your sequence converges to $0$.

btw, you don't really need the infinite series of $e^x$.

Try proving, by induction on $n$ that: $e^x \ge 1 + x + \frac{x^2}{2!} + \dots + \frac{x^n}{n!}$ for $x \ge 0$.

-

Expressing $n$ as $e^{\log n}$ The equation simplifies to $\text{exp}[{b \log n-na}]$

Since $\log n$ does not grow as fast as $n$, it should evaluate to zero as $n\rightarrow \infty$.

-

To me, this problem just screams for a certain Frenchman's aid:

Let $k$ be an integer greater than or equal to $b$. Then $$0\le{e^{-na}n^b}={n^b\over e^{na}}\le {n^k\over e^{na}}.$$ Now evaluate $\displaystyle\lim\limits_{n\rightarrow\infty}{ {n^k\over e^{na}}} =\lim\limits_{x\rightarrow\infty}{ {x^k\over e^{xa}}}$, by applying L'Hôpital's rule $k$-times.

(The first step isn't necessary, but in my opinion makes the write up a bit prettier.)

-
Nice, also thanks for using L'Hopital's rule, i needed a reminder of that. – Hardy Mar 6 '12 at 19:56