Proving that a sequence is within certain bounds

  • Thread starter Thread starter Elysian
  • Start date Start date
  • Tags Tags
    Bounds Sequence
Join the discussion
Ask a follow-up here, or get your own question answered by working scientists, mathematicians and engineers — people, not an autocomplete.
Real named experts · corrections over time · the nuance an AI answer skips
7 replies · 2K views
Elysian
Messages
33
Reaction score
0

Homework Statement


Define a1=1, and for every n>1, an+1 = an + [itex]\frac{1}{a<sub>n</sub>}[/itex]. Prove that 20 < a200 < 24

The Attempt at a Solution



I tried a few things to no avail. First, I showed that this is an increasing function by showing an+1 > an. I tried finding a limit, by saying if lim[itex]_{n\rightarrow∞}[/itex]an = L then lim[itex]_{n\rightarrow∞}[/itex]an+1 = L. Plugging in L, and solving but that didn't help. I feel like I need to find an upper and lower bound and work from there but I'm not sure how

Homework Statement


Homework Equations


The Attempt at a Solution

 
on Phys.org
Is this the complete question?
 
The first step is to get some idea of how fast an grows. If you think of the recurrence relation as a PDE, Δa = an+1 - an = 1/a then you can integrate to get a as a function of n. You'll then see how the n=200 relates to the lower bound of 20. Next, need to make up some inductive hypothesis for how far the sequence can deviate from the analytic solution.
 
haruspex said:
Hi Elysian... Have you solved this now, got too busy elsewhere, or given up? Did you understand my hint? I can guide you to the solution.

I did actually solve it, thank you for checking back. My bad on not replying. I found an upper and lower bound of the sequence, and with this I plugged in 200 into the lower bound of sqrt(2n), and the upper bound of sqrt(13n/6) and then got 20 and 20.866 respectively.
 
CompuChip said:
Of course 20.866 is certainly smaller than 24, but it seems a bit tight given the number in the question. This makes me a bit suspicious. Are you sure about your calculation? :-)

Pretty sure. To give you an overview of what I did,

I took an and looked at an+12. Which gave a formula an2 + 2 + [itex]\frac{1}{ a<sub>n+1</sub><sup>2</sup>}[/itex]. I compared an2 to 2n, and said that an2 > 2n for n[itex]\geq[/itex]3. Then by induction proved that this is the case for an+12 [itex]\geq[/itex] 2(n+1). Then I had to find a value greater than 2n, so I approximated that it would be (2+z)n, where z is a small number. Finding z to be 1/6, I plugged that in, got 13/6 as 2+z, and then continued the same process as before.