What Is the Sum of the Alternating Harmonic Series?

AI Thread Summary
The discussion focuses on finding the sum of the alternating harmonic series, expressed as Σ^{∞}_{k=1} (-1)^{k+1}/k. Participants confirm that the series converges due to the coefficients approaching zero and suggest using the Alternating Series Test for validation. While initial thoughts on telescoping and grouping terms were considered, it becomes clear that these methods are not suitable. The sum is ultimately identified as ln(2), and comparisons with known series are recommended for further understanding. The conversation emphasizes the importance of convergence and the relationship to Taylor series.
bdforbes
Messages
149
Reaction score
0
Find the sum of \Sigma^{\infty}_{k=1}\frac{(-1)^{k+1}}{k}

I know the series converges because the coefficients go monotonically to zero. However it's been a few years since I was taught how to sum these series, so I'm having trouble. I thought about telescoping, but no terms seem to cancel each other. I thought about maybe grouping terms together but I'm not sure how to proceed with that. I believe the sum should be ln(2), which implies those above methods aren't adequate. I was thinking maybe I need to compare the series with some other known series? A nudge in the right direction would be great, thanks.
 
Physics news on Phys.org
bdforbes said:
I need to compare the series with some other known series? A nudge in the right direction would be great, thanks.

That is one way of solving it. You already have one possible function to compare it against right in front of you.
 
Thanks you're right, I was being a bit stupid. I didn't actually have to find that sum anyway, just show that it converged, because then it would have to be equal to the value of the function whose Taylor series I was originally calculating.
That sentence made more sense in my mind.
 
bdforbes said:
Find the sum of \Sigma^{\infty}_{k=1}\frac{(-1)^{k+1}}{k}

I know the series converges because the coefficients go monotonically to zero.

I think the Alternating Series Test is pretty much what you would use here; this Test comes out of dealing with series like this. Things like the Comparison or Ratio Tests aren't much help: the Ratio Test gives the ambiguous result and the general term 1/k is already so simple that there isn't much to compare it against.

One thing you could do is a variant of the proof that the absolute harmonic series diverges. Consider the partial sums

1, 1 - (1/2), 1 - (1/2) + [(1/3) - (1/4)],
1 - (1/2) + [(1/3) - (1/4)] + [(1/5) - (1/6) + (1/7) - (1/8)], ...

versus 1, 1 + (1/2), 1 + (1/2) + (1/4), 1 + (1/2) + (1/4) + (1/8), ...

It's a little bit of work, but you can show that the next block of 2^(k-1) terms in the alternating harmonic series is smaller than the kth term in the geometric series which will converge to 2.

I thought about telescoping, but no terms seem to cancel each other.

I believe there that you're thinking of series like \sum_{k=1}^{N} \frac{1}{k} - \frac{1}{k+1}

And, yes, the sum does go to ln(2) -- see, for instance:

http://en.wikipedia.org/wiki/Harmonic_series_(mathematics )
http://mathworld.wolfram.com/AlternatingHarmonicSeries.html

This series also connects to some other interesting series.
 
Last edited by a moderator:
I tried to combine those 2 formulas but it didn't work. I tried using another case where there are 2 red balls and 2 blue balls only so when combining the formula I got ##\frac{(4-1)!}{2!2!}=\frac{3}{2}## which does not make sense. Is there any formula to calculate cyclic permutation of identical objects or I have to do it by listing all the possibilities? Thanks
Since ##px^9+q## is the factor, then ##x^9=\frac{-q}{p}## will be one of the roots. Let ##f(x)=27x^{18}+bx^9+70##, then: $$27\left(\frac{-q}{p}\right)^2+b\left(\frac{-q}{p}\right)+70=0$$ $$b=27 \frac{q}{p}+70 \frac{p}{q}$$ $$b=\frac{27q^2+70p^2}{pq}$$ From this expression, it looks like there is no greatest value of ##b## because increasing the value of ##p## and ##q## will also increase the value of ##b##. How to find the greatest value of ##b##? Thanks

Similar threads

Back
Top