Ask your own question, for FREE!
Mathematics 15 Online
OpenStudy (anonymous):

let a subscript 1 be a positive real number. Define a sequence a subscript n recursively by a subscript n +1 = a^2 subscript n -1. show that a subscript n does not converge to zero. Is there a subscript 1 such that the sequence a subscript n converges to some non-zero value?

OpenStudy (loser66):

wa ha, you have a big help. !!!

OpenStudy (anonymous):

$$a_n>0\\a_{n+1}=a_{n-1}^2$$is that what you're telling us?

OpenStudy (anonymous):

yes

OpenStudy (anonymous):

Well, presume our sequence converges to some non-zero value \(L\), i.e. \(\lim\limits_{n\to\infty}a_n=L\). Observe, then, that taking our recurrence relation in the limit gives:$$\lim_{n\to\infty}a_{n+1}=\lim_{n\to\infty}a_{n-1}^2\\L=L^2$$therefore either \(L=0\) or \(L=1\)

OpenStudy (anonymous):

err converges to some value \(L\) **

OpenStudy (anonymous):

We know that if \(a_1>1\) such a limit cannot exist, since then \(a_5>a_3>a_1\) etc. If \(a_1=1\) we observe that \(a_{2k+1}\) for integer \(k\) must be \(1\) and hence \(L=1\); for such a limit to exist we also require \(a_{2k}\to1\).

OpenStudy (anonymous):

If we let \(0<a_1<1\) we observe \(a_5<a_3<a_1\) etc. hence \(a_{2k+1}\to0\); since this limit corresponds to \(L=0\) we then infer that \(a_{2k}\to0\) as well.

OpenStudy (anonymous):

Intuitively then if \(a_1=1\) our sequence should converge to \(1\), which is nonzero.

OpenStudy (anonymous):

what was this part ("err converges to some value L **)?

OpenStudy (anonymous):

Note we took advantage of subsequence convergence...

OpenStudy (anonymous):

@mathtutoring22 well if we presume \(a_n\to L\) for some real value \(L\) we find our recurrence relation tells us \(L=L^2\) which has only two solutions, \(L=0\) and \(L=1\)

OpenStudy (anonymous):

I was correcting what I said earlier about \(L\) being nonzero

OpenStudy (anonymous):

ok, thanks.

OpenStudy (anonymous):

How do I prove that the limit ( as n goes to infinity) of 1/(nsin(n)) = 0? using l'hopital's or another rule

OpenStudy (anonymous):

$$\lim_{n\to\infty}\frac1{n\sin n}$$observe that since \(-1\le\sin n\le1\) we have \(-n\le n\sin n\le n\) for \(n\ge 0\). It follows that \(-\dfrac1n\ge\dfrac1{n\sin n}\ge\dfrac1n\). Now since in the limit as \(n\to\infty\) we have \(-\dfrac1n,\dfrac1n\to0\) so it follows that \(\dfrac1{n\sin n}\to 0\) -- this uses the 'squeeze' theorem

OpenStudy (anonymous):

wow. thanks. i tried using l'hospital's rule but i couldn't prove it.

OpenStudy (anonymous):

@mathtutoring22 are you sure it's \(a_{n+1}=(a_{n-1})^2\)... ? not, say, \(a_n=(a_{n-1})^2\) or \(a_{n+1}=a_n^2\)

OpenStudy (anonymous):

\[_{a _{n=1}} = ^{a^2_{n}} -1\]

OpenStudy (anonymous):

(a^2 subscript n) -1

OpenStudy (anonymous):

@oldrin.bataku

OpenStudy (anonymous):

OOOOOOOH

OpenStudy (anonymous):

Alright, so we have \(a_{n+1}=a_n^2-1\). Observe that if \(a_n\to L\) for some finite \(L\) in the limit as \(n\to\infty\) our recurrence relation tells us:$$\lim_{n\to\infty}a_{n+1}=\lim_{n\to\infty}(a_n^2-1)\\L=L^2-1\\L^2-L-1=0$$Ah, this problem is going to be fun ;-) can you recognize that polynomial?

OpenStudy (anonymous):

(x - (1+√5)/2) (x- (1-√5)/2)?

OpenStudy (anonymous):

well, one of its roots is \(\varphi\), the golden ratio, and its other is \(-\Phi\):$$L=\frac{1\pm\sqrt{1-4(-1)}}2=\frac{1\pm\sqrt5}2\\\varphi=\frac{1+\sqrt5}2,-\Phi=\frac{1-\sqrt5}2$$

OpenStudy (anonymous):

So clearly in the limit our value is either of \(\frac12(1\pm\sqrt5)\) if our sequence converges, and neither of those are \(0\), so we conclude our limit cannot conclude to \(0\)

OpenStudy (anonymous):

oops -- actually, if \(a_1<1\) the above occurs; if \(a_1=1\) our series becomes alternating and fails to converge: \(1,0,-1,0,-1,0,-1,0,-1,\dots\)

OpenStudy (anonymous):

and thats shows it?

OpenStudy (anonymous):

well i showed the first part now relax

OpenStudy (anonymous):

For the second part all we need to observe is that \(\varphi\) satisfies \(x=x^2-1\) (as we did above for \(L\)) so if we consider a constant sequence i.e. \(a_{n+1}=a_n\) we see that \(a_n=\varphi\) will give us a solution (so \(a_1=\varphi\)). If we could pick a negative \(a_1\), we could just use \(a_n=-\Phi\). http://www.wolframalpha.com/input/?i=a_1%3Dgolden+ratio%2C+a_%7Bn%2B1%7D%3Da_n%5E2+-+1 http://www.wolframalpha.com/input/?i=a_1%3Dconjugate+golden+ratio%2C+a_%7Bn%2B1%7D%3Da_n%5E2+-+1

OpenStudy (loser66):

@oldrin.bataku you make us like high school students. hehehe. .. which school did you graduate?

OpenStudy (anonymous):

@Loser66 I am a high school student :-p

OpenStudy (loser66):

if so, tell me what the school is, I turn down my degree, back there apply to be a high school student like you.

OpenStudy (anonymous):

I have two final question.

OpenStudy (anonymous):

the sequence \[_{a _{n}}\]

OpenStudy (anonymous):

is define in the following way: for every positive integer n we set a subscript n to be the sum of the positive divisors of n, for example a subscript 10 = 1+2+5+10 = 18. show that lim(as n goes to infinity) a subscript n / n is not = 2

OpenStudy (anonymous):

@oldrin.bataku

OpenStudy (anonymous):

this is my second and last question:

OpenStudy (anonymous):

@Loser66 haha I haven't learned this in school... but I have begun taking college classes as a high school student

OpenStudy (anonymous):

$$a_n=\sum d_k$$where \(d_k\) is the \(k\)-th positive divisor of \(n\).

OpenStudy (anonymous):

Well, first note we have \(a_p=p+1\) for all primes \(p\). Consider the infinite subsequence \(\{a_p\}\) of primes \(p\) and we see that since we have a infinitude of primes, \(p\) can grow arbitrarily large and thus \((p+1)/p=1+1/p\) will grow arbitrarily close to \(1\). Since thsi is not \(2\) we know \(\{a_n\}\) cannot converge to \(2\).

OpenStudy (anonymous):

Well consider that for \(n=1\) we have \((1+x)(1+x^2)=1+x+x^2+x^3\).

OpenStudy (anonymous):

Now, let's say it holds for some integer \(n\), i.e.$$(1+x)(1+x^2)\cdots(1+x^{2^n})=\sum_{k=0}^nx^k$$Consider that we can multiply by \((1+x^{2^{n+1}})\):$$(1+x)(1+x^2)\cdots(1+x^{2^{n+1}})=(1+x^{2^{n+1}})\sum_{k=0}^nx^k=\sum_{k=0}^nx^k+\sum_{k=0}^nx^{k+2^{n+1}}$$

OpenStudy (anonymous):

oops i messed up

OpenStudy (anonymous):

what the hell is \(m\)? from what I can tell we can let \(m=2^{n+1}-1\)

OpenStudy (anonymous):

Okay, let's rewrite:$$(1+x)(1+x^2)\cdots(1+x^{2^n})=\sum_{k=0}^mx^k\\(1+x)(1+x^2)\cdots(1+x^{2^{n+1}})=(1+x^{2^{n+1}})\sum_{k=0}^mx^k=\sum_{k=0}^m x^k+\sum_{k=0}^mx^{k+2^{n+1}}$$note we can rewrite our latter sum:$$\sum_{k=0}^mx^{k+2^{n+1}}=\sum_{k=2^{n+1}}^{m+2^{n+1}}x^k$$

OpenStudy (anonymous):

Since \(m=2^{n+1}-1\) we see our latter summation is really:$$\sum_{k=m+1}^{2m+1}x^k=\sum_{k=m+1}^{2^{n+2}-2+1}x^k=\sum_{k=m+1}^{2^{n+2}-1}x^k$$Combine our two sums:$$(1+x)(1+X^2)\cdots(1+x^{2^{n+1}})=\sum_{k=0}^{2^{n+2}-1}x^k$$and we see \(m_{n+1}=2^{n+1}-1\) giving us the form seen above; by the principle of induction we're done

OpenStudy (anonymous):

thanks again @oldrin.bataku @loser66

OpenStudy (loser66):

I did nothing, me helpless, friend.

OpenStudy (anonymous):

Where does the 2^n+1 comes from again, @oldrin.bataku @loser66

OpenStudy (anonymous):

@mathtutoring22 it's just the "m" that satisfies the prompt

OpenStudy (anonymous):

so replace 2^n+1 by m?

OpenStudy (anonymous):

oops i wrote that backwards

OpenStudy (anonymous):

What about this: prove that for every n>=, the number 11^n+1 + 12^2n-1 is divisible by 133. ( using math induction) I tried doing it but i made a mistake in the last part

Can't find your answer? Make a FREE account and ask your own questions, OR help others and earn volunteer hours!

Join our real-time social learning platform and learn together with your friends!
Can't find your answer? Make a FREE account and ask your own questions, OR help others and earn volunteer hours!

Join our real-time social learning platform and learn together with your friends!