Sequence

Moderators: Bogdan Posa, Laurian Filip, Beniamin Bogosel, Radu Titiu, Marius Dragoi

Post Reply
o.m.
Euclid
Posts: 32
Joined: Sun Apr 27, 2008 2:16 pm

Sequence

Post by o.m. »

x(1)>0

\( x(n+1)=\frac{x(n)}{1+n{x(n)}^2} \)

1/ Study the convergence of x(n).

2/ Find an equivalent of x(n).
bae
Bernoulli
Posts: 234
Joined: Tue Oct 02, 2007 10:39 pm

Post by bae »

***
Last edited by bae on Sat Feb 13, 2010 2:10 pm, edited 1 time in total.
aleph
Thales
Posts: 123
Joined: Mon Dec 24, 2007 2:06 am

Post by aleph »

Maple gives:
\( x(n) = {n}^{-1}+cn^{-2}+c^{2}n^{-3}+c^{3}n^{-4}+O\left( {n}^{-5}\right) \)
User avatar
Cezar Lupu
Site Admin
Posts: 612
Joined: Wed Sep 26, 2007 2:04 pm
Location: Bucuresti sau Constanta
Contact:

Re: Sequence

Post by Cezar Lupu »

o.m. wrote:x(1)>0

\( x(n+1)=\frac{x(n)}{1+n{x(n)}^2} \)

1/ Study the convergence of x(n).

2/ Find an equivalent of x(n).
For the first part:

1/ It is quite easy to see that from induction we get \( x_{n}>0, \forall n\geq 1 \). On the other hand, we have

\( \frac{x_{n+1}}{x_{n}}=\frac{1}{1+nx_{n}^{2}}<1, \)

so our sequence is decreasing, and by Weierstrass it is convergent.
But, from AM-GM inequality we get

\( x_{n+1}=\frac{x_{n}}{1+nx_{n}^{2}}\leq\frac{x_{n}}{2x_{n}\sqrt{n}}=\frac{1}{2\sqrt{n}} \).
Since \( x_{n}>0 \) by the sandwitch theorem it follows that \( \lim_{n\to\infty}x_{n}=0 \).
An infinite number of mathematicians walk into a bar. The first one orders a beer. The second orders half a beer. The third, a quarter of a beer. The bartender says “You’re all idiots”, and pours two beers.
Post Reply

Return to “Analiza matematica”