Intellectual battle with limits begins!

Calculus Level 2

L n = lim x ( ( x + a 1 ) ( x + a 2 ) ( x + a n ) ) 1 n x , L_n=\lim_{x\to\infty}((x+a_1)(x+a_2)\cdots(x+a_n))^{\frac{1}{n}}-x,

where a i = 1 2 i \large a_i = \frac{1}{2^i} .

Find lim n ( n L n ) \displaystyle{\lim_{n\to\infty} (nL_n)} .


The answer is 1.00.

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.

6 solutions

Ayush Verma
Oct 11, 2014

L e t x = 1 t s o , { ( x + a 1 ) ( x + a 2 ) . . . . ( x + a n ) } 1 n = { ( 1 + a 1 t ) ( 1 + a 2 t ) . . . ( 1 + a n t ) } 1 n t L e t y = { ( 1 + a 1 t ) ( 1 + a 2 t ) . . . ( 1 + a n t ) } 1 n L = l i m t 0 y 1 t = l i m t 0 { ( 1 + a 1 t ) ( 1 + a 2 t ) . . . ( 1 + a n t ) } 1 n 1 t ( 0 0 f o r m ) L = l i m t 0 y = l i m t 0 y n { a 1 1 + a 1 t + a 2 1 + a 2 t + . . . + a n 1 + a n t } n L = l i m t 0 { ( 1 + a 1 t ) ( 1 + a 2 t ) . . . ( 1 + a n t ) } 1 n { a 1 1 + a 1 t + a 2 1 + a 2 t + . . . + a n 1 + a n t } n L = r = 1 n a r l i m n n L = r = 1 a r = a 1 1 a 2 a 1 = 1 2 1 1 2 = 1 Let\quad x=\cfrac { 1 }{ t } \quad so,\\ \\ { \left\{ (x+{ a }_{ 1 })(x+{ a }_{ 2 })....(x+{ a }_{ n }) \right\} }^{ \cfrac { 1 }{ n } }=\cfrac { { { \left\{ (1+{ a }_{ 1 }t)(1+{ a }_{ 2 }t)...(1+{ a }_{ n }t) \right\} }^{ \cfrac { 1 }{ n } } } }{ t } \\ \\ Let\quad y={ \left\{ (1+{ a }_{ 1 }t)(1+{ a }_{ 2 }t)...(1+{ a }_{ n }t) \right\} }^{ \cfrac { 1 }{ n } }\\ \\ \Rightarrow L={ { lim }_{ t\rightarrow 0 }\cfrac { y-1 }{ t } =lim }_{ t\rightarrow 0 }\cfrac { { \left\{ (1+{ a }_{ 1 }t)(1+{ a }_{ 2 }t)...(1+{ a }_{ n }t) \right\} }^{ \cfrac { 1 }{ n } }-1 }{ t } \quad \quad (\cfrac { 0 }{ 0 } form)\\ \\ \Rightarrow L={ lim }_{ t\rightarrow 0 }{ y }^{ ' }={ lim }_{ t\rightarrow 0 }\cfrac { y }{ n } \left\{ \cfrac { { a }_{ 1 } }{ 1+{ a }_{ 1 }t } +\cfrac { { a }_{ 2 } }{ 1+{ a }_{ 2 }t } +...+\cfrac { { a }_{ n } }{ 1+{ a }_{ n }t } \right\} \\ \\ \Rightarrow nL={ lim }_{ t\rightarrow 0 }{ \left\{ (1+{ a }_{ 1 }t)(1+{ a }_{ 2 }t)...(1+{ a }_{ n }t) \right\} }^{ \cfrac { 1 }{ n } }\left\{ \cfrac { { a }_{ 1 } }{ 1+{ a }_{ 1 }t } +\cfrac { { a }_{ 2 } }{ 1+{ a }_{ 2 }t } +...+\cfrac { { a }_{ n } }{ 1+{ a }_{ n }t } \right\} \\ \\ \Rightarrow nL=\sum _{ r=1 }^{ n }{ { a }_{ r } } \\ \\ \Rightarrow { lim }_{ n\rightarrow \infty }nL=\sum _{ r=1 }^{ \infty }{ { a }_{ r } } =\cfrac { { a }_{ 1 } }{ 1-\cfrac { { a }_{ 2 } }{ { a }_{ 1 } } } =\cfrac { \cfrac { 1 }{ 2 } }{ 1-\cfrac { 1 }{ 2 } } =1

And this problem is lvl 2? Heh, what a joke XD

Trevor Arashiro - 5 years, 5 months ago

Log in to reply

At least lvl 4 I think...

展豪 張 - 5 years, 1 month ago

Log in to reply

@Trevor Arashiro because it becomes easy if solved with a "trick"

Ayush Verma - 4 years, 2 months ago

Could anyone tell me how to post a solution . I have one but unable to post it.

Sahil Jain - 4 years, 2 months ago

Log in to reply

if u have 'latex' problem,u can use Daum equation editor

Ayush Verma - 4 years, 2 months ago

Log in to reply

Does it work on android also.

Sahil Jain - 4 years, 2 months ago

Using latex start with "\ (" (no space) and end with "\ )". It's posted in the instructions somewhere on the full site.

Jerry McKenzie - 4 years, 1 month ago

Indeed. When 2 is changed into 1.9 or 2.1, the answer cannot be 1. Therefore, 1 2 1 1 2 = 1 \frac{\frac12}{1-\frac12} = 1 is correct.

Lu Chee Ket - 5 years, 5 months ago
Sandeep Bhardwaj
Oct 10, 2014

In this General Case :

L = lim x ( ( x + a 1 ) ( x + a 2 ) ( x + a n ) ) 1 n x = a 1 + a 2 + a 3 + . . + a n n L=\displaystyle \lim_{x \to \infty}((x+a_1)(x+a_2)••(x+a_n))^{\frac{1}{n}}-x= \dfrac{a_1+a_2+a_3+..+a_n}{n}

lim n n × ( a 1 + a 2 + a 3 + . . . . + a n n ) = a 1 + a 2 + a 3 + . . . . + a n \therefore \displaystyle \lim_{n\to\infty} n \times \left( \dfrac{a_1+a_2+a_3+....+a_n}{n} \right)=a_1+a_2+a_3+....+a_n

which is an infinite G.P.

R e q u i r e d L i m i t = ( a 1 1 a 2 a 1 ) = 1 \implies Required \ \ Limit=\left(\dfrac{a_1}{1-\frac{a_2}{a_1}}\right)=\boxed{1}

Forgot the "n" in the denominator of the first step,hehe.

pranav jangir - 5 years, 7 months ago

this is same as megh's method

Jaykant Shikre - 6 years, 8 months ago
Derek Khu
Aug 23, 2015

Consider x > 0 x > 0 and n > 0 n > 0 .

By the AM-GM-HM inequality, ( x + a 1 ) + + ( x + a n ) n x ( ( x + a 1 ) ( x + a n ) ) 1 n x n 1 x + a 1 + + 1 x + a n x \frac{(x+a_1)+ \cdots +(x+a_n)}{n}-x \geq ((x+a_1) \cdots (x+a_n))^{\frac{1}{n}}-x \geq \frac{n}{\frac{1}{x+a_1}+ \cdots + \frac{1}{x+a_n}} - x .

We have ( x + a 1 ) + + ( x + a n ) n x = a 1 + + a n n \frac{(x+a_1)+ \cdots +(x+a_n)}{n}-x = \frac{a_1 + \cdots + a_n}{n} .

Also,

n 1 x + a 1 + + 1 x + a n x = n ( x x + a 1 + + x x + a n ) 1 x + a 1 + + 1 x + a n = a 1 x + a 1 + + a n x + a n 1 x + a 1 + + 1 x + a n = ( a 1 + + a n ) x n 1 + O ( x n 2 ) n x n 1 + O ( x n 2 ) \begin{aligned} \frac{n}{\frac{1}{x+a_1}+ \cdots + \frac{1}{x+a_n}} - x & = \frac{n - (\frac{x}{x+a_1}+ \cdots + \frac{x}{x+a_n})}{\frac{1}{x+a_1}+ \cdots + \frac{1}{x+a_n}} \\ & = \frac{\frac{a_1}{x+a_1}+ \cdots + \frac{a_n}{x+a_n}}{\frac{1}{x+a_1}+ \cdots + \frac{1}{x+a_n}} \\ &= \frac{(a_1+ \cdots + a_n)x^{n-1} + O(x^{n-2})}{nx^{n-1} + O(x^{n-2})} \end{aligned}

where O ( x n 2 ) O(x^{n-2}) here represents a polynomial of degree at most n 2 n-2 . (In the last step, we multiplied both the numerator and denominator by ( x + a 1 ) ( x + a n ) (x+a_1) \cdots (x+a_n) .)

This gives us a 1 + + a n n ( ( x + a 1 ) ( x + a n ) ) 1 n x ( a 1 + + a n ) x n 1 + O ( x n 2 ) n x n 1 + O ( x n 2 ) \frac{a_1 + \cdots + a_n}{n} \geq ((x+a_1) \cdots (x+a_n))^{\frac{1}{n}}-x \geq \frac{(a_1+ \cdots + a_n)x^{n-1} + O(x^{n-2})}{nx^{n-1} + O(x^{n-2})} for all x > 0 x > 0 and n > 0 n > 0 . But as x x \to \infty , ( a 1 + + a n ) x n 1 + O ( x n 2 ) n x n 1 + O ( x n 2 ) a 1 + + a n n \frac{(a_1+ \cdots + a_n)x^{n-1} + O(x^{n-2})}{nx^{n-1} + O(x^{n-2})} \to \frac{a_1 + \cdots + a_n}{n} . So by sandwiching, ( ( x + a 1 ) ( x + a n ) ) 1 n x a 1 + + a n n ((x+a_1) \cdots (x+a_n))^{\frac{1}{n}}-x \to \frac{a_1 + \cdots + a_n}{n} as x x \to \infty , i.e. L L exists and equals a 1 + + a n n \frac{a_1 + \cdots + a_n}{n} .

Then n L = a 1 + + a n nL = a_1 + \cdots + a_n , and since the sequence ( a n ) (a_n) is a geometric progression with first term 1 2 \frac{1}{2} and common ratio 1 2 \frac{1}{2} , we know that lim n ( n L ) \lim_{n\to\infty} (nL) exists and equals 1 2 1 1 2 = 1 \frac{\frac{1}{2}}{1-\frac{1}{2}} = 1 .

Never realized how clever this solution was. Great method!

I never knew that AM=GM as the number of terms tends infinity if they're approximately equal.

Trevor Arashiro - 4 years, 2 months ago

Let's look at some element

x + a i x+a_i

You can write it as

x ( 1 + 1 x 2 i ) x (1+\frac{1}{x 2^i})

Using Bernoulli inequality that in limes becomes equality you have it's equal to

x ( 1 + 1 x ) 1 2 i x (1+\frac{1}{x})^{\frac{1}{2^i}}

Taking product of all of these we have it's equal to

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

The following sum is equal to 1 1

And now

n L = n ( ( x n ( 1 + 1 x ) ) 1 n x ) nL=n((x^n(1+\frac{1}{x}))^{\frac{1}{n}}-x)

n L = n ( x ( 1 + 1 n x ) x ) = 1 nL=n (x(1+\frac{1}{nx})-x)=1

Carsten Meyer
Feb 13, 2019

The solution can be done without de l'Hospital's formula for "0/0"-fractions, using a variant of the geometric sum. For simplicity, define

P n ( x ) : = k = 1 n ( x + a k ) = x n + x n 1 ( k = 1 n a k ) + = : x n + Q n 1 ( x ) L ~ n : = P n ( x ) 1 / n x \begin{aligned} P_n(x)&:=\prod_{k=1}^n(x+a_k)=x^n+x^{n-1}\left(\sum_{k=1}^na_k\right)+\ldots=:x^n+Q_{n-1}(x)\\\\ \tilde{L}_n&:=P_n(x)^{1/n}-x \end{aligned}

with polynomial degrees given by the index. Now to the variant of the geometric sum. Assuming we don't divide by zero, we have

a b = a n b n k = 0 n 1 a k b n 1 k , n N , a , b R \begin{aligned} a-b &= \frac{a^n-b^n}{ \sum_{k=0}^{n-1}a^kb^{n-1-k} },\qquad n\in\mathbb{N},\quad a,b\in\mathbb{R} \end{aligned}

Now consider the problem. We get rid of the n-th root and the subtraction by applying the above to L ~ n \tilde{L}_n :

L ~ n = P n ( x ) x n k = 0 n 1 P n ( x ) k n x n 1 k = Q n 1 ( x ) k = 0 n 1 P n ( x ) k n x n 1 k = Q n 1 ( x ) / x n 1 k = 0 n 1 ( P n ( x ) / x n ) k n x n + 1 x n + 1 L n = lim x L ~ n = k = 1 n a k k = 0 n 1 1 k n = 1 n k = 1 n a k , lim n ( n L n ) = lim n k = 1 n a k = k = 1 2 k = 1 q.e.d. \begin{aligned} \tilde{L}_n&=\frac{P_n(x)-x^n}{ \sum_{k=0}^{n-1}P_n(x)^{\frac{k}{n}}x^{n-1-k} }=\frac{ Q_{n-1}(x) }{ \sum_{k=0}^{n-1}P_n(x)^\frac{k}{n}x^{n-1-k} }=\frac{ Q_{n-1}(x)/x^{n-1} }{ \sum_{k=0}^{n-1}\left( P_n(x)/x^n \right)^{\frac{k}{n}} }\qquad\left|\cdot\frac{x^{-n+1}}{x^{-n+1}}\right.\\\\ \Rightarrow L_n&=\lim_{x\rightarrow\infty}\tilde{L}_n=\frac{\sum_{k=1}^na_k}{\sum_{k=0}^{n-1}1^{\frac{k}{n}}}=\frac{1}{n}\sum_{k=1}^na_k,\\\\ \lim_{n\rightarrow\infty}(nL_n)&=\lim_{n\rightarrow\infty}\sum_{k=1}^n a_k=\sum_{k=1}^\infty2^{-k}=1\qquad\text{q.e.d.} \end{aligned}

During the first limit x x\rightarrow\infty the first coefficient of Q n 1 ( x ) Q_{n-1}(x) remains in the numerator, while the denominator was normalized to 1.

Saúl Huerta
Aug 8, 2020

We can rewrite the limit as: L n = lim x x [ ( x n + O ( x n 1 ) x n ) 1 n 1 ] = lim x [ 1 + O ( 1 x ) ] 1 n 1 1 x L_n=\lim_{x\rightarrow\infty} x \left[\left(\dfrac{x^n+O(x^{n-1})}{x^n}\right)^{\frac{1}{n}}-1\right]=\lim_{x\rightarrow\infty} \dfrac{\left[1+O(\frac{1}{x})\right]^{\frac{1}{n}}-1}{\frac{1}{x}}

Where O ( x n 1 ) O(x^{n-1}) contains the rest of the terms of the polynomial and O ( 1 x ) O(\frac{1}{x}) contains these terms divided by x n x^n

Since it is of the 0 0 \frac{0}{0} form we can apply L'Hopital's Rule. L n = lim x 1 n [ 1 + O ( 1 x ) ] 1 n 1 [ O ( 1 x 2 ) ] 1 x 2 L_n=\lim_{x\rightarrow\infty} \dfrac{\frac{1}{n}\left[1+O(\frac{1}{x})\right]^{\frac{1}{n}-1}\left[O(\frac{1}{x^2})\right]}{-\frac{1}{x^2}}

Now we can multiply the denominator and numerator by x 2 -x^2 which will perfectly cancel the 1 x 2 -\frac{1}{x^2} in O ( 1 x 2 ) O(\frac{1}{x^2}) . The coefficient of the 1 x 2 -\dfrac{1}{x^2} term is j = 1 n 1 2 i \displaystyle\sum_{j=1}^{n}\dfrac{1}{2^i} and when multiplying x 2 -x^2 by the rest of the terms we are left with a bunch of x x terms in the denominators:

L n = 1 n j = 1 n 1 2 j \implies L_n=\dfrac{1}{n}\sum_{j=1}^{n}\dfrac{1}{2^j} lim n n L n = j = 1 1 2 j = 1 \lim_{n\rightarrow\infty} nL_n=\sum_{j=1}^{\infty}\dfrac{1}{2^j}=\boxed{1}

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...