21
Shubhodip
·2011-09-07 11:38:15
here is something within the syllabus
just recalled an well-known inequality 0\leq \frac{1}{2}.\frac{3}{4}.\frac{5}{6}\cdots\frac{2n-1}{2n}\leq\frac{1}{\sqrt{3n+1}} (we can prove this by induction)
using above
0\leq \frac{1}{2}.\frac{3}{4}.\frac{5}{6}\cdots\frac{2n-1}{2n}.\frac{2n+1}{2n+2}\leq \frac{1}{\sqrt{3n+4}}
taking \lim_{n\rightarrow\infty} we get what we want..
341
Hari Shankar
·2011-09-08 08:07:28
Here's one way at arriving at an upper bound for the expression (as shubhodip had already indicated):
http://www.goiit.com/posts/list/algebra-very-easyyyyy-75119.htm
62
Lokesh Verma
·2011-09-08 07:51:35
yup :|
*banging head on the wall*
1057
Ketan Chandak
·2011-09-08 06:13:45
sir yur statement proves the expression is > than 1n but as n→∞ 1n→0....so it proves that the expression is greater than 0.....
62
Lokesh Verma
·2011-09-08 04:17:40
read #9 .. it is very simple
1057
Ketan Chandak
·2011-09-08 04:14:56
but sir i cudnt have given a better answer keeping the fact in mind dat i havent learnt limits and asymptotic of central binomial coefficient nad stuffs like these.....but i think sir as it is a progressions sum there shud be a simpler way to solve the sum since no one learns limits before progressions....
62
Lokesh Verma
·2011-09-08 03:32:00
no it is not ketan.
both numerator and denominator tend to infinity.
but how are we sure that they relation is >>
and >> is a very relative term. it could mean a ratio of 1:10^100 and still not be equal to 0
"maths bhavna nahi jaanta" :P
1057
Ketan Chandak
·2011-09-07 21:59:42
sir,i know that the logic is not formal but what's wrong in my logic.....
as denomiator>>>numerator the fraction is tending to zero.....
62
Lokesh Verma
·2011-09-05 20:24:05
It is like this ketan
2[12(1-34)+1234(1-56)+................]
21
Shubhodip
·2011-09-07 10:46:24
here i finish the proof:( proof within syllabus should/may exist and i invite everyone to think of it,cz i am not doing the same right now )
I am going to use asymptotic of central binomial coefficient. We claim
\lim_{n\rightarrow\infty}\binom{2n}{n} = \sqrt{\frac{2}{\pi}}\frac{4^n}{\sqrt{2n+1}}
Here is an elementary proof.http://planetmath.org/encyclopedia/AsymptoticsOfCentralBinomialCoefficient.html This can be obtained by stirling's formula as well.
Follow 1st and 2nd line in my previous post (post 7).
The given expression , using above, becomes \lim_{n\rightarrow\infty}\frac{2\sqrt{2}}{\sqrt{\pi(2n+1)}} = 0
just learnt this
62
Lokesh Verma
·2011-09-07 09:57:38
Eureka?
12=12>
34>23
56>34
78>45
and so on
multiply both sides
u have LHS > 1/n :D
proved
62
Lokesh Verma
·2011-09-07 09:50:33
ketan.. that logic will not work!
I was wondering about a formal proof in the class but i could nto thik of one based on JEE syllabus.
infact now that i think even ratio test fails!
@subhodip! no clue buddy!
21
Shubhodip
·2011-09-07 09:42:23
may be :
given thing
= \lim_{n\rightarrow\infty}\frac{(2n+1)\binom{2n}{n}}{(n+1)4^n}
= \lim_{n\rightarrow\infty}\frac{2\binom{2n}{n}}{4^n}
=\lim_{n\rightarrow\infty}\frac{2\Gamma (n+\frac{1}{2})}{\sqrt{\pi}\Gamma (n+1)}
Isn't it acceptable that --
\lim_{n\rightarrow\infty}\frac{\Gamma (n+\frac{1}{2})}{\Gamma (n+1)}=0 ??
1057
Ketan Chandak
·2011-09-07 08:10:25
sir i think it is zero bcoz we find dat the denominator is always greater by 1 of each corresponding term in the numerator...so as the no terms is same as the n is very huge as it is tending to infinity,thus the denominator will be quite large as compared to the numerator....so the fraction will tend to zero.....
62
Lokesh Verma
·2011-09-07 06:52:19
A question which i have not been able to prove without root test!!
how are we srue that the thing u typed tends to zero?
1057
Ketan Chandak
·2011-09-06 05:12:43
it is like:
2[12(1-34)+1.32.4(1-56)+.......]
Breaking the brackets..
2[1.12-1.32.4+1.32.4........]
all the rest terms cancel out nd the last term will be 1.3.5........(2n-1)(2n+1)2.4.6.8.......2n(2n+2)
So Sn=2[12-1.3.5........(2n-1)(2n+1)2.4.6.8.......2n(2n+2)]
If n= ∞
1.3.5........(2n-1)(2n+1)2.4.6.8.......2n(2n+2) will be 0...
so Sn will be 1.....
11
Sambit Senapati
·2011-09-06 04:12:07
Ya but how to find the sum of the series?