How do you find the sum of an infinite series?

Click For Summary

Discussion Overview

The discussion revolves around finding the sum of the infinite series $$ \sum_{k=1}^{\infty} \frac{{k}^{2}+4}{{2}^{k}} $$, which is broken down into two parts: $$ \sum_{k=1}^{\infty} \frac{{k}^{2}}{{2}^{k}} $$ and $$ \sum_{k=1}^{\infty} \frac{4}{{2}^{k}} $$. Participants explore various methods to compute these sums, including partial sums, geometric series, and differentiation of power series.

Discussion Character

  • Exploratory
  • Mathematical reasoning
  • Debate/contested

Main Points Raised

  • Some participants express uncertainty about how to derive the sum from the term $$ \frac{n(n+1)(2n+1)}{6} $$ when using partial sums.
  • One participant notes that $$ \sum_{k=1}^{\infty} \frac{k^2}{2^k} $$ is not equal to $$ \sum_{k=1}^{n} \frac{k^2}{2^k} $$, emphasizing the distinction between the limit and the finite sum.
  • Another participant provides a method involving differentiation of a geometric series to derive the sum $$ \sum_{k=1}^\infty \frac{k^2}{2^k} $$, ultimately concluding it equals 6.
  • Some participants share their experiences with similar mathematical techniques and express admiration for innovative approaches used in the discussion.

Areas of Agreement / Disagreement

There is no consensus on the methods for calculating the sums, with multiple approaches presented and some participants expressing confusion or uncertainty about the steps involved. Disagreements exist regarding the validity of certain mathematical manipulations and interpretations of series convergence.

Contextual Notes

Participants mention limitations regarding the assumptions made in differentiating series and the conditions under which such operations are valid, particularly concerning the radius of convergence.

goohu
Messages
53
Reaction score
3
Hi, I'm trying to solve the sum of following infinite series:

$$ \sum_{k=1}^{\infty} \frac{{k}^{2}+4}{{2}^{k}} = \sum_{k=1}^{\infty} \frac{{k}^{2}}{{2}^{k}}
+ \sum_{k=1}^{\infty} \frac{4}{{2}^{k}}$$

Using partial sum we can rewrite the first series: $$ \sum_{k=1}^{\infty} \frac{{k}^{2}}{{2}^{k}} = \sum_{k=1}^{n} \frac{{k}^{2}}{{2}^{k}} = \frac{n(n+1)(2n+1)}{6} * \left(\frac{1}{2}\right)^{\!{k}} $$ , as n goes to $$\infty$$

The second series is a geometric one and can be rewritten as: $$ \sum_{k=1}^{\infty} \frac{4}{{2}^{k}} = 4* \sum_{k=1}^{\infty} \left(\frac{1}{2}\right)^{\!{k}} = 1/2 $$

It's unclear how to get the sum from the term $$ \frac{n(n+1)(2n+1)}{6} $$
 
Last edited:
Physics news on Phys.org
goohu said:
Hi, I'm trying to solve the sum of following infinite series:

$$ \sum_{k=1}^{\infty} \frac{{k}^{2}+4}{{2}^{k}} = \sum_{k=1}^{\infty} \frac{{k}^{2}}{{2}^{k}}
+ \sum_{k=1}^{\infty} \frac{4}{{2}^{k}}$$

Using partial sum we can rewrite the first series: $$ \sum_{k=1}^{\infty} \frac{{k}^{2}}{{2}^{k}} = \sum_{k=1}^{n} \frac{{k}^{2}}{{2}^{k}} = \frac{n(n+1)(2n+1)}{6} * \left(\frac{1}{2}\right)^{\!{k}} $$ , as n goes to $$\infty$$

The second series is a geometric one and can be rewritten as: $$ \sum_{k=1}^{\infty} \frac{4}{{2}^{k}} = 4* \sum_{k=1}^{\infty} \left(\frac{1}{2}\right)^{\!{k}} = 1/2 $$

It's unclear how to get the sum from the term $$ \frac{n(n+1)(2n+1)}{6} $$
A quick note:
[math]\sum_{k = 1}^{\infty} \dfrac{k^2}{2^k} \neq \sum_{k = 1}^{\infty} k^2 \cdot \sum_{k = 1}^{\infty} \dfrac{1}{2 ^k} [/math]

-Dan
 
Can I get hints? I've no idea how to continue from there
 
goohu said:
Hi, I'm trying to solve the sum of following infinite series:

$$ \sum_{k=1}^{\infty} \frac{{k}^{2}+4}{{2}^{k}} = \sum_{k=1}^{\infty} \frac{{k}^{2}}{{2}^{k}}
+ \sum_{k=1}^{\infty} \frac{4}{{2}^{k}}$$

Using partial sum we can rewrite the first series: $$ \sum_{k=1}^{\infty} \frac{{k}^{2}}{{2}^{k}} = \sum_{k=1}^{n} \frac{{k}^{2}}{{2}^{k}} = \frac{n(n+1)(2n+1)}{6} * \left(\frac{1}{2}\right)^{\!{k}} $$ , as n goes to $$\infty$$
I presume that last term was supposed to be [math]\left(\frac{1}{2}\right)^n[/math]. Also, [math]\sum_{k=1}^\infty
\frac{{k}^{2}}{{2}^{k}}[/math] is NOT equal to [math]\sum_{k=1}^n
\frac{{k}^{2}}{{2}^{k}}[/math]. The first is the limit of the latter as n goes to infinity.


The second series is a geometric one and can be rewritten as: $$ \sum_{k=1}^{\infty} \frac{4}{{2}^{k}} = 4* \sum_{k=1}^{\infty} \left(\frac{1}{2}\right)^{\!{k}} = 1/2 $$

It's unc+lear how to get the sum from the term $$ \frac{n(n+1)(2n+1)}{6} $$
 
$$\sum_{k=1}^\infty \frac{k^2+4}{2^k} = \sum_{k=1}^\infty \frac{k^2}{2^k} + \sum_{k=1}^\infty \frac{4}{2^k}$$

the last summation on the RHS is just the infinite geometric series ...

[math]\sum_{k=1}^\infty \frac{4}{2^k} = 4\sum_{k=1}^\infty \left(\frac{1}{2}\right)^k = 4 \cdot \frac{\frac{1}{2}}{1-\frac{1}{2}} = 4 \cdot 1 = 4[/math]for the first summation on the RHS (saw this done a long time ago, so bear with me) ...

$$\sum_{k=1}^\infty \frac{k^2}{2^k} = \frac{1}{2} + \frac{4}{4} + \frac{9}{8} + \frac{16}{16} + \frac{25}{32} + \,...$$

$$2 \sum_{k=1}^\infty \frac{k^2}{2^k} = 1 + \frac{4}{2} + \frac{9}{4} + \frac{16}{8} + \frac{25}{16} + \,...$$

$$2 \sum_{k=1}^\infty \frac{k^2}{2^k} - \sum_{k=1}^\infty \frac{k^2}{2^k} = 1 + \frac{3}{2} + \frac{5}{4} + \frac{7}{8} + \frac{9}{16} + \frac{11}{32} + \, ...$$

[math] \sum_{k=1}^\infty \frac{k^2}{2^k} - 1 = \frac{3}{2} + \frac{5}{4} + \frac{7}{8} + \frac{9}{16} + \frac{11}{32} + \, ...[/math]

[math] 2\left(\sum_{k=1}^\infty \frac{k^2}{2^k} - 1\right) = 3 + \frac{5}{2} + \frac{7}{4} + \frac{9}{8} + \frac{11}{16} + \, ...[/math]

[math] 2\left(\sum_{k=1}^\infty \frac{k^2}{2^k} - 1\right) - \left(2 \sum_{k=1}^\infty \frac{k^2}{2^k} - \sum_{k=1}^\infty \frac{k^2}{2^k}\right) = 2 + \frac{2}{2} + \frac{2}{4} + \frac{2}{8} + \frac{2}{16} + \, ...[/math]

[math]\sum_{k=1}^\infty \frac{k^2}{2^k} - 2 = 2 + \frac{2}{2} + \frac{2}{4} + \frac{2}{8} + \frac{2}{16} + \, ... = 2\left(1 +\frac{1}{2} + \frac{1}{4} + \frac{1}{8} + \frac{1}{16} + \, ... \right) = 2(2) = 4[/math]

[math]\sum_{k=1}^\infty \frac{k^2}{2^k} = 6[/math]
 
Nicely done! And very innovative.

I tried this one out on W|A but it wouldn't give the steps. It did come up with a partial sum that looks like it fell from outer space. I have no idea how the partial sum was got.

This reminds me of a number of high level Math proofs in Physics. First you define the problem. You work it out a ways. Then you redefine the problem. Rinse and repeat until you get something you know how to solve. (The combination of Green's functions and Fourier transforms still seems like the blackest of magics to me.)

-Dan
 
topsquark said:
Nicely done! And very innovative.

I tried this one out on W|A but it wouldn't give the steps. It did come up with a partial sum that looks like it fell from outer space. I have no idea how the partial sum was got.

This reminds me of a number of high level Math proofs in Physics. First you define the problem. You work it out a ways. Then you redefine the problem. Rinse and repeat until you get something you know how to solve. (The combination of Green's functions and Fourier transforms still seems like the blackest of magics to me.)

-Dan

I didn't come up with this on my own ... saw it at an AP Calculus workshop one summer where the presenter showed how to complete some difficult looking summations w/o calculus. Took me a while (and about 5-6 pages in my composition book) to recall the solution.
 
Another way to calculate the sum $$\sum_{k=1}^\infty \frac{k^2}{2^k}$$ is to start with the geometric series $$\sum_{k=1}^\infty \frac{x^k}{2^k} = \frac{\frac x2}{1-\frac x2} = \frac{x}{2-x}.$$ Differentiate both sides to get $$\sum_{k=1}^\infty \frac{kx^{k-1}}{2^k} = \frac2{(2-x)^2}.$$ Multiply both sides by $x$: $$\sum_{k=1}^\infty \frac{kx^k}{2^k} = \frac{2x}{(2-x)^2}.$$ Now differentiate again. After a bit of simplification, that becomes $$\sum_{k=1}^\infty \frac{k^2x^{k-1}}{2^k} = \frac{4+2x}{(2-x)^3}.$$ Finally, put $x=1$ and you get $$\sum_{k=1}^\infty \frac{k^2}{2^k} = 6.$$

In general, differentiating an infinite series of functions term by term is not a valid procedure. But there is a theorem which says that a power series can be differentiated term by term within its radius of convergence. The above series all have radius of convergence $2$, so the differentiation is valid in the region of $x=1$.
 

Similar threads

  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 3 ·
Replies
3
Views
4K
  • · Replies 5 ·
Replies
5
Views
3K
  • · Replies 15 ·
Replies
15
Views
3K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 4 ·
Replies
4
Views
3K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 9 ·
Replies
9
Views
3K
  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 5 ·
Replies
5
Views
3K