If you're seeing this message, it means we're having trouble loading external resources on our website.

If you're behind a web filter, please make sure that the domains *.kastatic.org and *.kasandbox.org are unblocked.

Main content

Worked example: divergent geometric series

AP.CALC:
LIM‑7 (EU)
,
LIM‑7.A (LO)
,
LIM‑7.A.3 (EK)
,
LIM‑7.A.4 (EK)
Sal evaluates the infinite geometric series -0.5+1.5-4.5+... Because the common ratio's absolute value is greater than 1, the series doesn't converge.

Want to join the conversation?

  • blobby green style avatar for user Rob
    Is there any way for us to know which infinity this series approaches (positive or negative)?
    (3 votes)
    Default Khan Academy avatar avatar for user
    • leafers seedling style avatar for user Shishir Iyer
      You don't. Since the numbers alternate between positive and negative, at some point, you will add a very large negative number to an even larger positive number to an even larger negative number. Because of this alternation between positive and negative for the partial sum, it cannot be determined whether it approaches an extremely large negative value or an extremely large positive value. If the common ratio were positive and greater than 1, though, it would approach positive infinity, since you are just adding up larger and larger positive numbers.
      (4 votes)
  • leafers ultimate style avatar for user Mamirov Yakhyo
    for the geometric series a+ar+ar^2+...
    the sum of the first two terms is 24 and the sum to infinity is 27.
    show that r=+-1/3. could you help, please
    (2 votes)
    Default Khan Academy avatar avatar for user
  • primosaur sapling style avatar for user Beanie
    does diverge mean you cannot add up the sum?
    (2 votes)
    Default Khan Academy avatar avatar for user
    • piceratops ultimate style avatar for user Rayan Murtada
      Basically, divergence means that the output value does not exist (could be positive or negative infinity, or does not approach a single non-imaginary value) as the limit approaches infinity.

      In the case with summation, adding an infinite number of terms will result in divergence when their sum is positive or negative infinite or does not approach a finite value.

      Conversely, convergence means that the output value does indeed exist (only one finite value). The sum of an infinite number of terms is convergent when adding an infinite terms approaches a single value.
      (3 votes)
  • orange juice squid orange style avatar for user vildan
    I cant get one point why "if |r|<1, it converges"? It's at 2.18. If you explain, i will appreciate.
    (3 votes)
    Default Khan Academy avatar avatar for user
    • hopper cool style avatar for user MonsterIsABlock
      This comes with the proof of the geometric series, which i will cover here.

      Imagine a series S going from 0 to a natural number n, inside of it is a constant a, multiplied by another constant r to the power of n, this means that we can write it as the following: a + ar^2 + ar^3 + ... ar^n-1, now if we multiply the series by r, we get ar + ar^2 + ar^3 + ... ar^n, now if we subtract S and rS we get: S - rS = S(1-r) = a + ar - ar + ar^2 - ar^2 .... ar^n-1 - ar^n-1 - ar^n, notice how all terms cancel except a and ar^n, so we get the following:

      S(1-r) = a - ar^n, divide both sides by 1-r we get:

      S = (a/1-r) - (ar^n/1-r), now if n is infinite, there are two possibilities, either r is less than 1 in which case the second term approaches zero and the infinite series converges, or r is more than 1 in which case, the 2nd term approaches infinity and the series diverges, i hope that this was clear.
      (1 vote)
  • piceratops seedling style avatar for user Brandon Park
    hello, I noticed that the geometric series started from n=0. Is there a reason why it doesn't start at 1? Personally, I think it might just be a matter of preference but I may be wrong. Thank you
    (1 vote)
    Default Khan Academy avatar avatar for user
    • leaf green style avatar for user cossine
      n is the number of terms.
      Having the power of series from 0 to n-1 mean there are n terms because:

      (n-1) - 0 + 1 = n

      Think of the following question:
      How many number are there from 0 to 150:

      Answer: 150 - 0 + 1 = 151

      Since we need to include 0 we have plus 1. Arguably you could say why not have the powers from 1 to n. In this case the first term is ar, however in my experience things become a bit confusing.

      E.g. for the sequence 2, 4, 8 etc this will cause a = 1 instead of 2 since r = 2 => ar = 2a = 2 => a=1 and .

      So I would recommend the power of a geometric series are defined from 0 to n-1.
      (2 votes)
  • blobby green style avatar for user NAKOJA K. JEREMIAH
    what does it mean to say a series is bounded or is not bounded
    (1 vote)
    Default Khan Academy avatar avatar for user
    • hopper cool style avatar for user Iron Programming
      If you have a series F, and have another series L; then if series F is always less than series L for all terms n, then series F is bounded by series L because it can't ever be greater than it. This is useful for proving divergence or convergence in series.

      Hope this helps,
      - Convenient Colleague
      (1 vote)
  • mr pants green style avatar for user Annanickoloff
    Does the fact that the series does not converge necessarily mean that the series diverges or does it just mean that the test is inconclusive?
    (1 vote)
    Default Khan Academy avatar avatar for user
  • duskpin seedling style avatar for user Jasmine979
    what happens when n=1?
    (1 vote)
    Default Khan Academy avatar avatar for user
    • duskpin ultimate style avatar for user Jennifer Andrey
      The n=0 under the sigma sign just indicates that for the first term in this series, you plug in 0 for n in the formula:
      (-0.5)(-3)^0
      Because (-3) to the zero power is just 1, your first term is (-0.5)(1) or -0.5.
      If n=1, the first term in the series would have to be when you plug in 1 for n in the formula:
      (-0.5)(-3)^1
      Which gives us 1.5.
      This is of course the second term in the first series, where we were given n=0. N is just the starting value, and tells us what the first term of the series is going to be. In this case, the series are the same, and the value of n just tells us which term we start with. For n=0, it is -0.5, and for n=1, it is 1.5.
      I hope this answered your question.
      (1 vote)
  • primosaur tree style avatar for user Hanna
    I don't really get this methodology of divergent and convergence thing? I just don't get the concept of what and why we are calculating a limit, and what does it even mean to "Converge" on 12, huh?
    (0 votes)
    Default Khan Academy avatar avatar for user
    • leaf green style avatar for user ( ͡° ͜ʖ ͡°)
      As you add more and more terms to your sum, each term can either be big enough that it makes the sum approach infinity as the number of terms increases, or it can be small enough that it makes the sum approach a specific number (not infinity) as the number of terms increases. A divergent sum is one that gets bigger and bigger as you add more terms, and a convergent sum is one that "converges" or gets closer and closer to a number (if it converges on 12, it just means that if you add an infinite number of infinitely tiny terms in the series, it all adds up to 12). For a sum to be convergent, it both has to approach zero as the number of terms approaches infinity (which is why you use a limit to check if this is true), and it also has to pass some different tests we can use to analyze series.
      (2 votes)

Video transcript

- [Voiceover] So we've got this infinite series here, and let's see, it looks like a geometric series. We're gonna go from this term to the second term, we are multiplying by negative three, and then to go to the next term we're gonna multiply by negative three again. So it looks like we have a common ratio of negative three. So we could actually rewrite this series as being equal to negative 0.5, I could say times negative three to the zero power, negative three to the zero power, plus negative zero, or maybe I can just keep writing this way, minus 0.5 times negative three to the first power, times negative three to the first power, minus 0.5, minus 0.5 times negative three to the second power, negative three to the second power. And we're just gonna keep going like that. And we could just say we're just gonna keep having minus 0.5 times negative three to each or to a higher and higher and higher powers. Or we could write this in sigma notation. This is equal to the same thing, as the sum from, let's say n equals zero to infinity, we're just gonna keep going on and on forever. And it's gonna be this first, it's gonna be, you can kind of think the thing we're multiplying by negative three to some power, so it's gonna be negative 0.5, actually, I'm just gonna do that yellow color. So it's gonna be negative 0.5 times negative three, negative, I'm gonna do that blue color, so times negative three to the nth power. Here this is when n is zero, here's n is one, here is n is equal to two. So we've been able to rewrite this in different ways, but let's actually see if we can evaluate this. And so we have a common ratio of negative three. So our r here is negative three. And the first thing that you should think about is well, in order for this to converge, our common ratio, the magnitude of the common ratio, or the absolute value of the common ratio needs to be less than one for convergence. And what is the absolute value of negative three? Well, the absolute value of negative three is equal to three which is definitely not less than one. So this thing will not converge. This thing will not converge. And even if you look at this, it makes sense, 'cause the magnitudes of each of these terms are getting larger and larger and larger. We're flipping between adding and subtracting. But we're adding and subtracting larger and larger and larger and larger and larger values. Intuitively, when things converge you're kind of, each successive term tends to get diminishingly small or maybe it cancels out in some type of an interesting way. But because the absolute value of the common ratio is greater than or equal to one in this situation, this is not going to converge to a value.