# Thread: Past ACTM question. Limits.

1. ## Past ACTM question. Limits.

$\lim_{x\rightarrow\infty}\left (\frac{1+2^{199}+3^{199}+4^{199}...+n^{199}}{n^{20 0}}\right )$

Is equal to:
a. 1/201
b.1/200
c.1/199
d.1/98
e. None of these

My attempt:
I have no clue how to take this limit so I just tried to use calculator guessing.
$\frac{1}{1^{200}}=1$

$\frac{1+2^{199}}{2^{200}}=.5$

$\frac{1+2^{199}+3^{199}}{3^{200}}=.3333...$

$\frac{1+2^{199}+3^{199}+4^{199}}{4^{200}}=.25...$

And my calculator wont go past this so I just figured that it continued with the pattern that it has now.

so
keeping with the pattern:
if n=1
it's
1^{-1}
if n=2
it's
2^{-1}
if n=3
it's
3^{-1} ect.
it must be zero because 1/infinity=0
so e

The answer however is B. Does anyone know how to work this limit?

2. The method I used is recognizing the limit is actually a definite integral (in the form of a Riemann sum). (I wouldn't be surprised if someone else posted an easier solution involving a subtle algebraic manipulation of the actual limit instead of converting it to an integral).

Maybe you remember that a definite integral is defined as a Riemann sum. If not, then here is the definition of a definite integral:

$\displaystyle \int_a^b f(x) \, dx = \lim_{n\to\infty} \sum_{i=1}^n f(x_i) \Delta x_i$

For the rest of this post, I am going to show that your problem is actually a Riemann sum (the right-hand side of the above equation). By doing so, I am allowed to convert it to an equivalent definite integral, which will hopefully be easy to evaluate!

If the question were to find the area between f(x) = x^199 and the x-axis on the interval [a,b] using a Riemann sum with n rectangles, what would you do? The Riemann sum would look like:

$\displaystyle \lim_{n\to\infty} \sum_{i=1}^n f(x_i) \Delta x_i$.

If we decide each of our rectangles will have the same width, then

$\Delta x_i = \Delta x = \frac{b-a}{n}$.

And then, $x_i = a + \Delta x \cdot i = a + \frac{b-a}{n} i$.

Substituting these values into the Riemann sum gives:

$\displaystyle \lim_{n\to\infty} \sum_{i=1}^n \left [ \left ( a + \frac{b-a}{n} i \right ) ^{199} \cdot \frac{b-a}{n} \right ]$

This looks nothing like the original problem, right? Well, the first term of our summation is (i = 1):

$(a + \frac{b-a}{n})^{199} \cdot \frac{b - a}{n}$.

If we look back at the original problem, we can imagine the first term being $\frac{1}{n^{200}}$, which is equal to $\left ( \frac{1}{n} \right ) ^{199} \cdot \frac{1}{n}$. They are starting to look similar (a little!).

Now, the natural question is what values of a and b will make the first term in the Riemann sum and the first term in the original problem match exactly?

You may already notice that $(a + \frac{b-a}{n})^{199}$ will probably have to match $\left ( \frac{1}{n} \right ) ^{199}$. Similarly, $\frac{b-a}{n}$ will probably have to match $\frac{1}{n}$. Let's focus on $(a + \frac{b-a}{n})^{199}$ and $\left ( \frac{1}{n} \right ) ^{199}$.

$\left ( \frac{1}{n} \right ) ^{199}$ doesn't have any addition. It isn't $\left (10 + \frac{1}{n} \right ) ^{199}$, it's just $\left ( \frac{1}{n} \right ) ^{199}$. If a = 0, then $(a + \frac{b-a}{n})^{199}$ becomes $\left ( \frac{1}{n} \right ) ^{199}$. They match!

Now, let's switch our focus to $\frac{b-a}{n}$ and $\frac{1}{n}$. We now know a = 0, so:

$\frac{b}{n}$ will have to match $\frac{1}{n}$.

We can conclude b = 1. Substituting a and b into the Riemann sum and checking other terms, such as i = 2 or i = 3, shows that our Riemann sum now matches the original problem exactly. In other words, I have just shown that your problem is actually a Riemann sum with a = 0 and b = 1. According to the definition of a definite integral, a Riemann sum is just a definite integral. Therefore, your problem is a definite integral!

Now we know the original problem is equivalent to evaluating $\int_a^b x^{199} \, dx$, where a = 0 and b = 1.

$\displaystyle \lim_{n\to\infty} \dfrac{1+2^{199}+3^{199}+4^{199}...+n^{199}}{n^{20 0}} = \int_0^1 x^{199} \, dx = \frac{x^{200}}{200} \bigg |_0^1 = \frac{1}{200}$

Like I said, I wouldn't be surprised if someone now posts a much more elegant solution that has eluded me. Until then, this is the best I've got.

In summary, the original problem is actually a definite integral. Once the limits of integration were found, it was only a matter of evaluating an integral.

Hi, Jose. <3

$\displaystyle \int_a^b f(x) \, dx = \lim_{n\to\infty} \sum_{i=1}^n f(x_i) \Delta x_i$