A long limit

Calculus Level 5

The value of the limit

L = lim y ( 8 y + 3 ) ( y + 1 ) ( y + 2 ) ( y + 3 ) ( 2 y + 3 ) ( 2 y + 5 ) ( 4 y + 5 ) 7 2 y L=\displaystyle\lim_{y \rightarrow \infty} \sqrt[7]{(8y+3)(y+1)(y+2)(y+3)(2y+3)(2y+5)(4y+5)}-2y

is of the form A B \dfrac{A}{B} where A A and B B are co-prime integers.

Find the value of A + B A+B .

Note : \textbf{Note :} Do NOT \textbf{NOT} use wolfram alpha.


The answer is 121.

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.

5 solutions

Patrick Corn
Jul 21, 2014

Let f ( y ) f(y) be that seventh root. We have that f ( y ) 2 y = f ( y ) 7 128 y 7 f ( y ) 6 + ( 2 y ) f ( y ) 5 + + ( 2 y ) 6 . f(y) - 2y = \frac{f(y)^7-128y^7}{f(y)^6 + (2y)f(y)^5 + \cdots + (2y)^6}. The top of this fraction is a polynomial of degree 6 6 , equal to 1488 y 6 + c 5 y 5 + + c 0 1488 y^6 + c_5 y^5 + \cdots + c_0 . Divide the top and bottom by y 6 y^6 to get 1488 + c 5 / y + + c 0 / y 6 ( f ( y ) / y ) 6 + 2 ( f ( y ) / y ) 5 + + 2 6 . \frac{1488 + c_5/y + \cdots + c_0/y^6}{(f(y)/y)^6 + 2(f(y)/y)^5 + \cdots + 2^6}.

Since lim y f ( y ) y = 2 \lim\limits_{y\to\infty} \frac{f(y)}{y} = 2 , the limit we want is 1488 2 6 + 2 6 + 2 6 = 1488 448 = 93 28 \frac{1488}{2^6 + 2^6 + \cdots 2^6} = \frac{1488}{448} = \frac{93}{28} . So the answer is 121 \fbox{121} .

Heh, I tried to find some sort of relationship to AM-GM (seeing 7 terms multiplied to the one seventh power).

Finn Hulse - 6 years, 10 months ago
Michael Mendrin
Jul 21, 2014

The slope of the asymptote of the radical

( 8 y + 3 ) ( y + 1 ) ( y + 2 ) ( y + 3 ) ( 2 y + 3 ) ( 2 y + 5 ) ( 4 y + 5 ) 7 \sqrt [ 7 ]{ (8y+3)(y+1)(y+2)(y+3)(2y+3)(2y+5)(4y+5) }

is 2 y 2y , because the highest order term in the radical (when expanded) is 128 y 7 128{ y }^{ 7 } . Hence, this is cancelled by the subtraction of 2 y 2y from the radical, leaving a limit value which is the same as the y y intercept of the asymptote. Let a a be that y y intercept. Then the coefficient of the 2 n d 2nd term of both the expanded polynomial in the radical and ( 2 y + a ) 7 { (2y+a) }^{ 7 } has to be the same. Thus a a has to be

a = 128 ( 3 8 + 1 1 + 2 1 + 3 1 + 5 2 + 5 4 ) 128 ( 7 ) ( 1 2 ) = 93 28 = A B a=\dfrac { 128(\dfrac { 3 }{ 8 } +\dfrac { 1 }{ 1 } +\dfrac { 2 }{ 1 } +\dfrac { 3 }{ 1 } +\dfrac { 5 }{ 2 } +\dfrac { 5 }{ 4 } ) }{ 128(7)(\dfrac { 1 }{ 2 } ) } =\dfrac { 93 }{ 28 } =\dfrac { A }{ B }

Pratik Shastri
Jul 19, 2014

A weird solution ----

We need to find

L = lim y ( 8 y + 3 ) ( y + 1 ) ( y + 2 ) ( y + 3 ) ( 2 y + 3 ) ( 2 y + 5 ) ( 4 y + 5 ) 7 2 y L=\displaystyle\lim_{y \rightarrow \infty} \sqrt[7]{(8y+3)(y+1)(y+2)(y+3)(2y+3)(2y+5)(4y+5)}-2y .

We can rewrite it as

L = lim y 2 × ( y + 3 8 ) ( y + 1 ) ( y + 2 ) ( y + 3 ) ( y + 3 2 ) ( y + 5 2 ) ( y + 5 4 ) 7 2 y L=\displaystyle\lim_{y \rightarrow \infty} 2 \times \sqrt[7]{\left(y+\frac{3}{8}\right)(y+1)(y+2)(y+3)\left(y+\frac{3}{2}\right)\left(y+\frac{5}{2}\right)\left(y+\frac{5}{4}\right)}-2y

Now, consider the terms ( y + 3 8 ) , ( y + 1 ) , ( y + 2 ) , ( y + 3 ) , ( y + 3 2 ) , ( y + 5 2 ) \left(y+\frac{3}{8}\right), (y+1), (y+2), (y+3), \left(y+\frac{3}{2}\right), \left(y+\frac{5}{2}\right) and ( y + 5 4 ) \left(y+\frac{5}{4}\right) .

As y y \rightarrow \infty , all these terms become equal.

We can check this by dividing each of the above terms by every other term and then applying lim y \displaystyle\lim_{y \rightarrow \infty} to it.

Hence, we can apply the fact that for equal terms, A . M = G . M A.M=G.M .

Doing so,

lim y ( y + 3 8 ) ( y + 1 ) ( y + 2 ) ( y + 3 ) ( y + 3 2 ) ( y + 5 2 ) ( y + 5 4 ) 7 = lim y ( y + 3 8 ) + ( y + 1 ) + ( y + 2 ) + ( y + 3 ) + ( y + 3 2 ) + ( y + 5 2 ) + ( y + 5 4 ) 7 \displaystyle\lim_{y \rightarrow \infty} \sqrt[7]{\left(y+\frac{3}{8}\right)(y+1)(y+2)(y+3)\left(y+\frac{3}{2}\right)\left(y+\frac{5}{2}\right)\left(y+\frac{5}{4}\right)}=\displaystyle\lim_{y \rightarrow \infty} \dfrac{\left(y+\frac{3}{8}\right)+(y+1)+(y+2)+(y+3)+\left(y+\frac{3}{2}\right)+\left(y+\frac{5}{2}\right)+\left(y+\frac{5}{4}\right)}{7} .

Hence,

L = lim y 2 ( 93 56 + y ) 2 y L= \displaystyle\lim_{y \rightarrow \infty} 2\left(\dfrac{93}{56}+y\right)-2y

L = 93 28 \boxed{L=\dfrac{93}{28}}

A + B = 121 \therefore \boxed{A+B=121}

Your argument is not mathematically correct. You have the right idea, but need to figure out how to show the limit properly.

Calvin Lin Staff - 6 years, 10 months ago

Log in to reply

What's the mistake in this method??..i have also solved this question by using this method.... Please tell me

Kïñshük Sïñgh - 6 years, 10 months ago

Log in to reply

The statement of "As y y \rightarrow \infty , all these terms become equal. Hence we can apply the fact that for equal terms AM= GM." is false.

For example, we know that lim 2 y \lim 2y and lim y 2 \lim \frac{ y}{2} also approach the same limit (namely infinity). Does this mean that

lim 2 y × y 2 = lim 2 y + y 2 2 = 5 4 y ? \lim \sqrt{ 2y \times \frac{ y}{2} } = \lim \frac{ 2y + \frac{y}{2} } { 2} = \frac{ 5}{4} y ?

This approach "happens to work" for "linear factors with the same coefficient". As such, you need a proper explanation for that fact.

Calvin Lin Staff - 6 years, 10 months ago

Log in to reply

@Calvin Lin Can't we say that in comparison to y other terms like 3..... can be neglected
So AM=GM. Also in your example Lim \frac{2×y}{y÷2} =4 not 1 Please reply and correct me if I am wrong

AYUSH JAIN - 5 years ago

Log in to reply

@Ayush Jain As I said

This approach "happens to work" for "linear factors with the same coefficient". As such, you need a proper explanation for that fact.

My example isn't perfectly equivalent. So, what is the mathematical reasoning for why it works in the scenario that you're thinking of? Is "limit of ratios = 1" sufficient?

Calvin Lin Staff - 5 years ago

Log in to reply

@Calvin Lin Let us take x=10,000. Now 2x=20,000
Whereas x+3=10,003 now what I am trying to say through this is if we multiply something to a big number it increases or decreases appreciably but if we add a small number to it it does not change much . That is why AM=GM method is working only where coeficient of x=1. Calvin Lin sir please reply and help me to improve myself

AYUSH JAIN - 5 years ago

Log in to reply

@Ayush Jain

What is the mathematical reasoning for why it works in the scenario that you're thinking of? Is "limit of ratios = 1" sufficient?

Calvin Lin Staff - 5 years ago

Log in to reply

@Calvin Lin I think necessary condition are: X should tend to infinity Coefficient of x should be 1 The term added with x should be finite. Please reply. Thanks

AYUSH JAIN - 5 years ago

You can check that with the following limit(s)--

L = lim y ( y + 3 8 y + 1 ) = 1 L=\displaystyle\lim_{y \rightarrow \infty} \left(\dfrac{y+\frac{3}{8}}{y+1}\right)=1

And similarly for every other term.

Pratik Shastri - 6 years, 10 months ago

Log in to reply

Yes, the limit of the ratio is equal to 1 is the relevant property here. Now, how do you use this to justify the interchange of limits?

Calvin Lin Staff - 6 years, 10 months ago
David Vaccaro
Jul 23, 2014

Take out factor of 2 y 2y and use Binomial Expansion on each of the sevenths roots.

Solving the problem by using CAUCHY's INEQUALITY:

L = lim y [ ( 8 y + 3 ) ( y + 1 ) ( y + 2 ) ( y + 3 ) ( 2 y + 3 ) ( 2 y + 5 ) ( 4 y + 5 ) 7 2 y ] = 2 lim y [ ( y + 3 / 8 ) ( y + 1 ) ( y + 2 ) ( y + 3 ) ( y + 3 / 2 ) ( y + 5 / 2 ) ( y + 5 / 4 ) 7 y ] L\quad =\quad \lim _{ y\rightarrow \infty }{ [\sqrt [ 7 ]{ (8y+3)(y+1)(y+2)(y+3)(2y+3)(2y+5)(4y+5) } -2y] } \quad =\quad \\ 2\lim _{ y\rightarrow \infty }{ [\sqrt [ 7 ]{ (y+3/8)(y+1)(y+2)(y+3)(y+3/2)(y+5/2)(y+5/4) } -y } ]

by Cauchy's inequality:

( y + 3 / 8 ) ( y + 1 ) ( y + 2 ) ( y + 3 ) ( y + 3 / 2 ) ( y + 5 / 2 ) ( y + 5 / 4 ) 7 ( y + 3 / 8 ) + ( y + 1 ) + ( y + 2 ) + ( y + 3 ) + ( y + 3 / 2 ) + ( y + 5 / 2 ) + ( y + 5 / 4 ) 7 = y + 93 56 \sqrt [ 7 ]{ (y+3/8)(y+1)(y+2)(y+3)(y+3/2)(y+5/2)(y+5/4) } \le \\ \le \quad \frac { (y+3/8)+(y+1)+(y+2)+(y+3)+(y+3/2)+(y+5/2)+(y+5/4) }{ 7 } =y+\frac { 93 }{ 56 }

where equality holds for y + 3 / 8 y + 1 y + 2 y + 3 y + 3 / 2 y + 5 / 2 y + 5 / 4 y+3/8\approx y+1\approx y+2\approx y+3\approx y+3/2\approx y+5/2\approx y+5/4 it works if y y\rightarrow \infty

lim y [ ( y + 3 / 8 ) ( y + 1 ) ( y + 2 ) ( y + 3 ) ( y + 3 / 2 ) ( y + 5 / 2 ) ( y + 5 / 4 ) 7 = y + 93 56 \Longrightarrow \lim _{ y\rightarrow \infty }{ [\sqrt [ 7 ]{ (y+3/8)(y+1)(y+2)(y+3)(y+3/2)(y+5/2)(y+5/4) } \quad =\quad } y+\frac { 93 }{ 56 }

L = 2 ( y + 93 56 y ) = 93 28 = A B A n s w e r i s A + B = 121 \Longrightarrow \quad L\quad =\quad 2(y+\frac { 93 }{ 56 } -y)=\frac { 93 }{ 28 } =\frac { A }{ B } \quad \quad \quad \quad \Longrightarrow \quad Answer\quad is\quad A+B=121

Note that you have only shown that it is a valid upper limit. The argument that this is the least upper bound needs to be more rigorous.

Calvin Lin Staff - 6 years, 10 months ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...