Inspired by Rohit Ner

Calculus Level 5

lim N n = 1 N ( ( 1 ) n + 1 ( 2 k = 0 2 n ( 1 ) k ( 2 n k ) ) ) \displaystyle\lim_{N \rightarrow \infty} \sum_{n=1}^{N} \left( (-1)^{n+1} \left( 2 - \sum_{k=0}^{2n} \dfrac{(-1)^{k}}{\dbinom{2n}{k}} \right) \right)

If the above expression equals a ln ( b ) , a - \ln(b), where a a and b b are positive integers, then find a 3 + b 3 . a^{3} + b^{3}.


The answer is 9.

This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try refreshing the page, (b) enabling javascript if it is disabled on your browser and, finally, (c) loading the non-javascript version of this page . We're sorry about the hassle.

2 solutions

Yong See Foo
Jun 8, 2015

First we prove the following identity which we shall use later, using the formula for the sum of terms in a geometric progression: k = 0 2 n ( 1 t ) 2 n ( t t 1 ) k = ( 1 t ) 2 n ( 1 ( t t 1 ) 2 n + 1 ) 1 t t 1 = ( 1 t ) 2 n + 1 + t 2 n + 1 \begin{aligned} \sum_{k=0}^{2n} (1-t)^{2n}\left(\frac{t}{t-1}\right)^k &=(1-t)^{2n} \frac{(1-\left(\frac{t}{t-1}\right)^{2n+1})}{1-\frac{t}{t-1}} \\ &=(1-t)^{2n+1}+t^{2n+1} \end{aligned} (the last step involved splitting the numerator and taking out the 1 -1 in the second term, while simplifying the denominator). Now we utilise Euler's well-known Beta function: B ( x , y ) = ( x 1 ) ! ( y 1 ) ! ) ( x + y 1 ) ! = 0 1 t x 1 ( 1 t ) y 1 d t . B(x,y)=\frac{(x-1)!(y-1)!)}{(x+y-1)!}=\int_0^1 t^{x-1}(1-t)^{y-1} dt. Putting what we have until here together - replacing ( x 1 , y 1 ) (x-1,y-1) with ( k , n k ) (k,n-k) tells us k = 0 2 n ( 1 ) k ( 2 n k ) = k = 0 2 n ( 1 ) k k ! ( 2 n k ) ! ( 2 n + 1 ) ( 2 n + 1 ) ! = k = 0 2 n ( 1 ) k ( 2 n + 1 ) 0 1 t k ( 1 t ) 2 n k d t = ( 2 n + 1 ) 0 1 k = 0 2 n ( 1 t ) 2 n ( t t 1 ) k d t = ( 2 n + 1 ) 0 1 ( 1 t ) 2 n + 1 d t + 0 1 t 2 n + 1 d t = 2 ( 2 n + 1 ) 0 1 t 2 n + 1 d t = 2 n + 1 n + 1 \begin{aligned} \sum_{k=0}^{2n}\frac{(-1)^k}{{2n \choose k}} &= \sum_{k=0}^{2n}\frac{(-1)^kk!(2n-k)!(2n+1)}{(2n+1)!} \\ &= \sum_{k=0}^{2n}(-1)^k(2n+1)\int_0^1 t^{k}(1-t)^{2n-k} dt\\ &= (2n+1)\int_0^1\sum_{k=0}^{2n} (1-t)^{2n}\left(\frac{t}{t-1}\right)^kdt\\ &= (2n+1)\int_0^1(1-t)^{2n+1}dt+\int_0^1t^{2n+1}dt \\ &=2(2n+1)\int_0^1t^{2n+1}dt=\frac{2n+1}{n+1} \end{aligned} by using the identity before to get rid of the sum. For the sake of completeness, I shall also show that the alternating sum of the harmonic series converges to ln ( 2 ) \text{ln}(2) . Note that n = 1 ( 1 ) n + 1 n = n = 0 ( 1 ) n n + 1 = n = 0 0 1 ( x ) n d x = 0 1 n = 0 1 x + 1 d x = ln ( 2 ) \begin{aligned} \sum_{n=1}^\infty \frac{(-1)^{n+1}}{n} &= \sum_{n=0}^\infty \frac{(-1)^n}{n+1} \\ &= \sum_{n=0}^\infty\int_0^1 (-x)^n dx\\ &=\int_0^1 \sum_{n=0}^\infty\frac{1}{x+1} dx\\ &=\text{ln}(2) \end{aligned} by using the representation of the sum of an infinite geometric progression which is valid as we are still within the radius of convergence. Now although Brian has presented rest of the solution, I will just complete it, we know now that the expression which limit we want to find is n = 1 n ( 1 ) n + 1 ( 2 2 n + 1 n + 1 ) = n = 1 n ( 1 ) n + 1 1 n + 1 \sum_{n=1}^n (-1)^{n+1}\left(2-\frac{2n+1}{n+1}\right)=\sum_{n=1}^n (-1)^{n+1}\frac{1}{n+1} and hence the limit is 1 n = 1 n ( 1 ) n + 1 n = 1 ln ( 2 ) , 1-\sum_{n=1}^n \frac{(-1)^{n+1}}{n}=1-\text{ln}(2), giving a final answer of 1 3 + 2 3 = 9 1^3+2^3=\boxed{9} .

Great solution! Thanks for posting it. :)

Brian Charlesworth - 6 years ago

I'll expand this solution later, but for the time being, note that

2 k = 0 2 n ( 1 ) k ( 2 n k ) = 2 2 ( 2 n + 1 ) 2 n + 2 = 1 n + 1 . 2 - \displaystyle\sum_{k=0}^{2n} \dfrac{(-1)^{k}}{\dbinom{2n}{k}} = 2 - \dfrac{2(2n + 1)}{2n + 2} = \dfrac{1}{n + 1}.

(See here for a discussion on this type of sum.)

Then lim N n = 1 N ( 1 ) n + 1 n + 1 = 1 ln ( 2 ) . \lim_{N \rightarrow \infty} \displaystyle\sum_{n=1}^{N} \dfrac{(-1)^{n+1}}{n + 1} = 1 - \ln(2).

Thus a 3 + b 3 = 1 3 + 2 3 = 9 . a^{3} + b^{3} = 1^{3} + 2^{3} = \boxed{9}.

Hello sir,

I want to discuss about a problem I m working on...is there any way I can contact to you(I need your help)?

Krishna Sharma - 6 years ago

Log in to reply

We can discuss the problem on this page if you'd like. There doesn't seem to be much interest in this question anymore so few if any would see the discussion.

Brian Charlesworth - 6 years ago

Log in to reply

Ok....this is only a part of the problem which I have to drop it because I cannot solve it but then I got interested in it...so here it is

A given circle rolls on a given ellipse then the maximum distance of the point of contact from the line joining centres is?

This came to my mind because at axes the distance will be zero hence there will be a point in between where the distance will be maximum...soon after doubting myself I started thinking that is point of contact always lie on line joining centres?

Please answer this and don't get angry if you find this silly.

Krishna Sharma - 6 years ago

Log in to reply

@Krishna Sharma I've thought about this same situation before and never got around to working out the details. The point of contact is not on the line joining the centers except for the four extrema, (unless the ellipse is a circle, in which case the point of contact always lies on this line), so there will indeed be some maximum distance as you describe. My initial cursory attempt at finding this maximum was a mess, so I will need to give this more thought when I have time. This is tough and definitely not "silly". :)

Brian Charlesworth - 6 years ago

Log in to reply

@Brian Charlesworth Thanks for replying :D

Wow, this is tough for you....then I'll definitely try this. Some gif could be very helpful for visualization of the situation...

Krishna Sharma - 6 years ago

Log in to reply

@Krishna Sharma Sorry for the delay in getting back to you. I haven't as yet found any better approach than the one based on this link .

Since the distance from the point P P to the line joining the centers is c sin ( z ) c*\sin(\angle z) the question essentially becomes one of maximizing z \angle z for the given parameters of the ellipse and circle. As you can see the final solution would not be a simple one, and would likely rely heavily on WolframAlpha (or the like).

This seems like such a natural question to ask that I'm surprised that there aren't more references to it.

Brian Charlesworth - 6 years ago

Log in to reply

@Brian Charlesworth No problem :D Yes I have seen that link before but in that problem additional information is given about the coordinates of centre of circle.

I tried this problem made 2 nasty equations and what I got in the end...1=1 -_- What I'm feeling is that information is insufficient to solve the problem on paper.

I'll be posting few ellipse related problems and your special circle fitting also :p

Krishna Sharma - 6 years ago

Log in to reply

@Krishna Sharma Haha. Well, at least you didn't end up with 1 = 2. :) I look forward to trying out your ellipse problems; maybe I'll try to come up with one of my own as well.

Brian Charlesworth - 6 years ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...