Difference between revisions of "2018 AMC 12A Problems/Problem 21"

m (Solution 2.5 (simpler, inequalities))
(Video solution by grogg007 (5 mins, simple solution))
 
(21 intermediate revisions by 2 users not shown)
Line 35: Line 35:
 
~cpma213
 
~cpma213
  
==Solution 2.5 (inequalities)==  
+
==Solution 3 (Simpler Inequalities)==  
Our real roots must be in the interval <math>(-1, 0).</math> Immediately, we see the root of choice E is around <math>-1</math> which is the smallest possible value on this interval, so we can safely eliminate this.  
+
The real roots must lie in the interval <math>(-1, 0).</math> The root of choice E is around <math>-1</math> which is the smallest value on this interval, so we can safely eliminate this.
 +
 
 +
Let's pick a number between <math>0</math> and <math>-1,</math> say <math>-1/2.</math> We'll plug this in for <math>x.</math> Note that plugging in <math>-1/2</math> here will give us positive values for all polynomials. This means their roots are less than <math>-1/2</math> since the graph increases from <math>x = -1</math> to <math>0.</math> Therefore, the polynomials giving the smallest values when <math>-1/2</math> is plugged in for <math>x</math> have roots closer to <math>-1/2</math> (closer to the positive end of the interval) which means those roots are larger.
  
Let's pick a number between <math>0</math> and <math>-1,</math> say <math>-1/2.</math> We'll plug this in for <math>x.</math> Note that plugging in <math>-1/2</math> here will give us positive values for all polynomials. This means their roots are less than <math>-1/2.</math> Therefore, the polynomials giving the smallest values when <math>-1/2</math> is plugged in for <math>x</math> have roots closer to <math>-1/2,</math> which makes those roots larger.
 
  
 
Comparing choices A and B:
 
Comparing choices A and B:
Line 44: Line 45:
 
<cmath>(-1/2)^{19} + 2018(-1/2)^{11} + 1 > (-1/2)^{17} + 2018(-1/2)^{11} + 1</cmath>
 
<cmath>(-1/2)^{19} + 2018(-1/2)^{11} + 1 > (-1/2)^{17} + 2018(-1/2)^{11} + 1</cmath>
  
Comparing choices C and D:  
+
Comparing choices C and D:
  
 
<cmath>(-1/2)^{19} + 2018(-1/2)^{13} + 1 > (-1/2)^{17} + 2018(-1/2)^{13} + 1</cmath>
 
<cmath>(-1/2)^{19} + 2018(-1/2)^{13} + 1 > (-1/2)^{17} + 2018(-1/2)^{13} + 1</cmath>
Line 52: Line 53:
  
 
<cmath>(-1/2)^{17} + 2018(-1/2)^{13} + 1 > (-1/2)^{17} + 2018(-1/2)^{11} + 1</cmath>
 
<cmath>(-1/2)^{17} + 2018(-1/2)^{13} + 1 > (-1/2)^{17} + 2018(-1/2)^{11} + 1</cmath>
 +
  
 
Since choice B gave a smaller value than choice D, <math>\boxed{\textbf{(B) } x^{17}+2018x^{11}+1}</math> has the largest roots.
 
Since choice B gave a smaller value than choice D, <math>\boxed{\textbf{(B) } x^{17}+2018x^{11}+1}</math> has the largest roots.
Line 57: Line 59:
 
~[[User:grogg007|grogg007]]
 
~[[User:grogg007|grogg007]]
  
Desmos: https://www.desmos.com/calculator/bgibeqbiku
+
Desmos (you’ll need to zoom in): https://www.desmos.com/calculator/bgibeqbiku
  
== Solution 3 (Similar to Solution 1) ==
+
== Solution 4 (Similar to Solution 1) ==
 
Let the real solution to <math>B</math> be <math>a.</math> It is easy to see that when <math>a</math> is plugged in to <math>A,</math> since <math>-1 < a < 0,</math> it follows that <math>a^{19} < a^{17}</math> thus making the real solution to <math>A</math> more "negative", or smaller than <math>B.</math> Similarly we can assert that <math>D > C.</math> Now to compare <math>B</math> and <math>D,</math> we can use the same method to what we used before to compare <math>B</math> to <math>A,</math> in which it is easy to see that the smaller exponent <math>(11)</math> "wins". Now, the only thing left is for us to compare <math>B</math> and <math>E.</math> Plugging <math>\frac{-2018}{2019}</math> (or the solution to <math>E</math>) into <math>B</math> we obtain <math>\frac{(-2018)^{17}}{2019^{17}} + 2018\cdot\frac{(-2018)^{11}}{2019^{11}} + 1,</math> which is intuitively close to <math>-1 - 2018 + 1 = -2018, </math> much smaller than the solution the required <math>0.</math> (For a more rigorous proof, one can note that <math>\left(\frac{2018}{2019}\right)^{17}</math> and <math>\left(\frac{2018}{2019}\right)^{11}</math> are both much greater than <math>\left(\frac{2018}{2019}\right)^{2019} \approx \frac{1}{e},</math> by the limit definition of <math>e.</math> Since <math>- \frac{1}{e} - 2018 \cdot \frac{1}{e} + 1</math> is still much smaller the required <math>0</math> for the solution to <math>B</math> to be a solution, our answer is <math>\boxed{\textbf{(B) }  x^{17}+2018x^{11}+1}.</math>
 
Let the real solution to <math>B</math> be <math>a.</math> It is easy to see that when <math>a</math> is plugged in to <math>A,</math> since <math>-1 < a < 0,</math> it follows that <math>a^{19} < a^{17}</math> thus making the real solution to <math>A</math> more "negative", or smaller than <math>B.</math> Similarly we can assert that <math>D > C.</math> Now to compare <math>B</math> and <math>D,</math> we can use the same method to what we used before to compare <math>B</math> to <math>A,</math> in which it is easy to see that the smaller exponent <math>(11)</math> "wins". Now, the only thing left is for us to compare <math>B</math> and <math>E.</math> Plugging <math>\frac{-2018}{2019}</math> (or the solution to <math>E</math>) into <math>B</math> we obtain <math>\frac{(-2018)^{17}}{2019^{17}} + 2018\cdot\frac{(-2018)^{11}}{2019^{11}} + 1,</math> which is intuitively close to <math>-1 - 2018 + 1 = -2018, </math> much smaller than the solution the required <math>0.</math> (For a more rigorous proof, one can note that <math>\left(\frac{2018}{2019}\right)^{17}</math> and <math>\left(\frac{2018}{2019}\right)^{11}</math> are both much greater than <math>\left(\frac{2018}{2019}\right)^{2019} \approx \frac{1}{e},</math> by the limit definition of <math>e.</math> Since <math>- \frac{1}{e} - 2018 \cdot \frac{1}{e} + 1</math> is still much smaller the required <math>0</math> for the solution to <math>B</math> to be a solution, our answer is <math>\boxed{\textbf{(B) }  x^{17}+2018x^{11}+1}.</math>
  
 
~fidgetboss_4000
 
~fidgetboss_4000
  
==Solution 4==
+
==Solution 5==
  
 
Denote the polynomials in the answer choices by <math>A(x),B(x),C(x),D(x),</math> and <math>E(x),</math> respectively.
 
Denote the polynomials in the answer choices by <math>A(x),B(x),C(x),D(x),</math> and <math>E(x),</math> respectively.
Line 83: Line 85:
 
~[https://artofproblemsolving.com/wiki/index.php/User:Isabelchen isabelchen]
 
~[https://artofproblemsolving.com/wiki/index.php/User:Isabelchen isabelchen]
  
==Solution 5 (Monotonic Function)==
+
==Solution 6 (Monotonic Function)==
  
 
Denote the polynomials in the answer choices by <math>A(x),B(x),C(x),D(x),</math> and <math>E(x),</math> respectively.
 
Denote the polynomials in the answer choices by <math>A(x),B(x),C(x),D(x),</math> and <math>E(x),</math> respectively.
Line 111: Line 113:
 
~[https://artofproblemsolving.com/wiki/index.php/User:Isabelchen isabelchen]
 
~[https://artofproblemsolving.com/wiki/index.php/User:Isabelchen isabelchen]
  
==Solution 6 (Proof by Contradiction)==
+
==Solution 7 (Proof by Contradiction)==
  
 
Denote the polynomials in the answer choices by <math>A(x),B(x),C(x),D(x),</math> and <math>E(x),</math> respectively.
 
Denote the polynomials in the answer choices by <math>A(x),B(x),C(x),D(x),</math> and <math>E(x),</math> respectively.
Line 156: Line 158:
 
~[https://artofproblemsolving.com/wiki/index.php/User:Isabelchen isabelchen]
 
~[https://artofproblemsolving.com/wiki/index.php/User:Isabelchen isabelchen]
  
== Solution 7 (Calculus) ==
+
== Solution 8 (Calculus) ==
 
Note that <math>a(-1)=b(-1)=c(-1)=d(-1) < 0</math> and <math>a(0)=b(0)=c(0)=d(0) > 0</math>. Calculating the definite integral for each function in the interval <math>[-1,0]</math>, we see that <math>B(x)\rvert^{0}_{-1}</math> gives the most negative value. To maximize our real root, we want to maximize the area of the curve under the x-axis, which means we want our integral to be as negative as possible and thus the answer is <math>\boxed{\textbf{(B) }  x^{17}+2018x^{11}+1}</math>.
 
Note that <math>a(-1)=b(-1)=c(-1)=d(-1) < 0</math> and <math>a(0)=b(0)=c(0)=d(0) > 0</math>. Calculating the definite integral for each function in the interval <math>[-1,0]</math>, we see that <math>B(x)\rvert^{0}_{-1}</math> gives the most negative value. To maximize our real root, we want to maximize the area of the curve under the x-axis, which means we want our integral to be as negative as possible and thus the answer is <math>\boxed{\textbf{(B) }  x^{17}+2018x^{11}+1}</math>.
  
== Solution 8 (Calculus) ==
+
== Solution 9 (Calculus) ==
 
Newton's Method is used to approximate the zero <math>x_{1}</math> of any real valued function given an estimation for the root <math>x_{0}: \ x_{1}=x_{0}-{\frac {f(x_{0})}{f'(x_{0})}}.</math> After looking at all the options, <math>x_{0}=-1</math> gives a reasonable estimate. For options <math>\textbf{(A)}</math> to <math>\textbf{(D)},</math> we have <math>f(-1) = -2018</math> and the estimation becomes <math>x_{1}=-1+{\frac {2018}{f'(-1)}}.</math> Thus we need to minimize the derivative, giving us <math>\textbf{(B)}</math>. Now after comparing <math>\textbf{(B)}</math> and <math>\textbf{(E)}</math> through Newton's method, we see that <math>\textbf{(B)}</math> has the higher root, so the answer is <math>\boxed{\textbf{(B) }  x^{17}+2018x^{11}+1}</math>.
 
Newton's Method is used to approximate the zero <math>x_{1}</math> of any real valued function given an estimation for the root <math>x_{0}: \ x_{1}=x_{0}-{\frac {f(x_{0})}{f'(x_{0})}}.</math> After looking at all the options, <math>x_{0}=-1</math> gives a reasonable estimate. For options <math>\textbf{(A)}</math> to <math>\textbf{(D)},</math> we have <math>f(-1) = -2018</math> and the estimation becomes <math>x_{1}=-1+{\frac {2018}{f'(-1)}}.</math> Thus we need to minimize the derivative, giving us <math>\textbf{(B)}</math>. Now after comparing <math>\textbf{(B)}</math> and <math>\textbf{(E)}</math> through Newton's method, we see that <math>\textbf{(B)}</math> has the higher root, so the answer is <math>\boxed{\textbf{(B) }  x^{17}+2018x^{11}+1}</math>.
  
 
~Qcumber
 
~Qcumber
 +
 +
== Solution 10 (very fast) ==
 +
Obviously, eliminate <math>(E).</math> Note that for answer choices <math>(A)</math> and <math>(B),</math> <math>x \approx \sqrt[11]{\left(-\frac{1}{2018}\right)} > \sqrt[13]{\left(-\frac{1}{2018}\right)},</math> with the RHS being approximately the value for <math>x</math> when considering answer choices <math>(C)</math> or <math>(D),</math> so the answer is either <math>(A)</math> or <math>(B).</math> Now, since <math>x^{19} < x^{17}</math> for all negative reals <math>x</math> (we want to choose the option closer to <math>0</math> because then we will have to decrease <math>x</math> less since we estimated), we have <math>\boxed{(B)}</math> as the clear answer.
 +
 +
~ martianrunner
  
 
== Video Solution by Richard Rusczyk ==
 
== Video Solution by Richard Rusczyk ==
Line 168: Line 175:
  
 
~ dolphin7
 
~ dolphin7
 +
 +
==Video Solution by [[User:grogg007|grogg007]] (5 mins, simple solution)==
 +
https://www.youtube.com/watch?v=kjjCdK8lPDY
  
 
==See Also==
 
==See Also==
 
{{AMC12 box|year=2018|ab=A|num-b=20|num-a=22}}
 
{{AMC12 box|year=2018|ab=A|num-b=20|num-a=22}}
 
{{MAA Notice}}
 
{{MAA Notice}}

Latest revision as of 13:07, 11 August 2025

Problem

Which of the following polynomials has the greatest real root?

$\textbf{(A) }   x^{19}+2018x^{11}+1   \qquad        \textbf{(B) }   x^{17}+2018x^{11}+1   \qquad    \textbf{(C) }   x^{19}+2018x^{13}+1   \qquad   \textbf{(D) }  x^{17}+2018x^{13}+1 \qquad  \textbf{(E) }   2019x+2018$

Solution 1 (Intermediate Value Theorem, Inequalities, Graphs)

Denote the polynomials in the answer choices by $A(x),B(x),C(x),D(x),$ and $E(x),$ respectively.

Note that $A(x),B(x),C(x),D(x),$ and $E(x)$ are strictly increasing functions with range $(-\infty,\infty).$ So, each polynomial has exactly one real root. The real root of $E(x)$ is $x=-\frac{2018}{2019}\approx-1.000.$ On the other hand, since $A(-1)=B(-1)=C(-1)=D(-1)=-2018$ and $A(0)=B(0)=C(0)=D(0)=1,$ we conclude that the real root for each of $A(x),B(x),C(x),$ and $D(x)$ must satisfy $x\in(-1,0)$ by the Intermediate Value Theorem (IVT).

We analyze the polynomials for $x\in(-1,0):$

  1. We have \begin{align*} B(x)-A(x)=D(x)-C(x)&=x^{17}-x^{19} \\ &=x^{17}\left(1-x^2\right) \\ &<0. \end{align*} As the graph of $y=A(x)$ is always above the graph of $y=B(x)$ in this interval, we deduce that $B(x)$ has a greater real root than $A(x)$ does. By the same reasoning, $D(x)$ has a greater real root than $C(x)$ does.
  2. We have \begin{align*} B(x)-D(x)&=2018x^{11}-2018x^{13} \\ &=2018x^{11}\left(1-x^2\right) \\ &<0, \end{align*} from which $B(x)$ has a greater real root than $D(x)$ does.

Now, we are left with comparing the real roots of $B(x)$ and $E(x).$ Since $B\left(-\frac{1}{\sqrt2}\right)<0<B(0),$ it follows that the real root of $B(x)$ must satisfy $x\in\left(-\frac{1}{\sqrt2},0\right)$ by the IVT. Clearly, $\boxed{\textbf{(B) }   x^{17}+2018x^{11}+1}$ has the greatest real root.

~MRENTHUSIASM

Solution 2 (Similar to Solution 1)

We can see that our real solution has to lie in the open interval $(-1,0)$. From there, note that $x^a < x^b$ if $a$, $b$ are odd positive integers if $a<b$, so hence it can only either be $\textbf{(B)}$ or $\textbf{(E)}$ (as all of the other polynomials will be larger than the polynomial $\textbf{(B)}$). Observe that $\textbf{(E)}$ gives the solution $x=-\frac{2018}{2019}$. We can approximate the root for $\textbf{(B)}$ by using $x=-\frac 12$: \[\left(-\frac{1}{2}\right)^{17} - \frac{2018}{2048} + 1 \approx 0.\] Therefore, the root for $\textbf{(B)}$ is approximately $-\frac 12$. The answer is $\boxed{\textbf{(B) }   x^{17}+2018x^{11}+1}$.

~cpma213

Solution 3 (Simpler Inequalities)

The real roots must lie in the interval $(-1, 0).$ The root of choice E is around $-1$ which is the smallest value on this interval, so we can safely eliminate this.

Let's pick a number between $0$ and $-1,$ say $-1/2.$ We'll plug this in for $x.$ Note that plugging in $-1/2$ here will give us positive values for all polynomials. This means their roots are less than $-1/2$ since the graph increases from $x = -1$ to $0.$ Therefore, the polynomials giving the smallest values when $-1/2$ is plugged in for $x$ have roots closer to $-1/2$ (closer to the positive end of the interval) which means those roots are larger.


Comparing choices A and B:

\[(-1/2)^{19} + 2018(-1/2)^{11} + 1 > (-1/2)^{17} + 2018(-1/2)^{11} + 1\]

Comparing choices C and D:

\[(-1/2)^{19} + 2018(-1/2)^{13} + 1 > (-1/2)^{17} + 2018(-1/2)^{13} + 1\]


We're left with choices B and D since they gave the smallest values:

\[(-1/2)^{17} + 2018(-1/2)^{13} + 1 > (-1/2)^{17} + 2018(-1/2)^{11} + 1\]


Since choice B gave a smaller value than choice D, $\boxed{\textbf{(B) } x^{17}+2018x^{11}+1}$ has the largest roots.

~grogg007

Desmos (you’ll need to zoom in): https://www.desmos.com/calculator/bgibeqbiku

Solution 4 (Similar to Solution 1)

Let the real solution to $B$ be $a.$ It is easy to see that when $a$ is plugged in to $A,$ since $-1 < a < 0,$ it follows that $a^{19} < a^{17}$ thus making the real solution to $A$ more "negative", or smaller than $B.$ Similarly we can assert that $D > C.$ Now to compare $B$ and $D,$ we can use the same method to what we used before to compare $B$ to $A,$ in which it is easy to see that the smaller exponent $(11)$ "wins". Now, the only thing left is for us to compare $B$ and $E.$ Plugging $\frac{-2018}{2019}$ (or the solution to $E$) into $B$ we obtain $\frac{(-2018)^{17}}{2019^{17}} + 2018\cdot\frac{(-2018)^{11}}{2019^{11}} + 1,$ which is intuitively close to $-1 - 2018 + 1 = -2018,$ much smaller than the solution the required $0.$ (For a more rigorous proof, one can note that $\left(\frac{2018}{2019}\right)^{17}$ and $\left(\frac{2018}{2019}\right)^{11}$ are both much greater than $\left(\frac{2018}{2019}\right)^{2019} \approx \frac{1}{e},$ by the limit definition of $e.$ Since $- \frac{1}{e} - 2018 \cdot \frac{1}{e} + 1$ is still much smaller the required $0$ for the solution to $B$ to be a solution, our answer is $\boxed{\textbf{(B) }   x^{17}+2018x^{11}+1}.$

~fidgetboss_4000

Solution 5

Denote the polynomials in the answer choices by $A(x),B(x),C(x),D(x),$ and $E(x),$ respectively.

The real root of $E(x)$ is $x=-\frac{2018}{2019}\approx-1.000.$, therefore, we can eliminate this polynomial first. Notice that all of the real roots for all answer choices are in $(-1,0)$. Therefore, \[x^{11}<0, \quad x^2<1, \quad x^{13}>x^{11}, \quad x^{19}+2018x^{13}+1>x^{19}+2018x^{11}+1, \quad C(x)>A(x)\] \[\quad x^{17}+2018x^{13}+1>x^{17}+2018x^{11}+1, \quad D(x)>B(x)\] \[x^{17}<0, \quad x^{19}>x^{17}, \quad x^{19}+2018x^{11}+1>x^{17}+2018x^{11}+1, \quad A(x)>B(x)\]

Now compare the real roots of $A(x)$ and $D(x)$. \[A(x) - D(x) = x^{19}+2018x^{11}+1 - x^{17}-2018x^{13}-1 = x^{17}(x^2-1) - 2018x^{11}(x^2-1) = x^{11}(x^2-1)(x^6-2018)\] \[\because x^{11}<0, \quad 0<x^2<1, \quad 0<x^6<1, \quad \therefore A(x) - D(x)<0, \quad A(x) < D(x)\]

\[C(x)>D(x)>A(x)>B(x)\]

As $B(x)$ has the smallest value for the same $x$, and $A(x),B(x),C(x),D(x),$ are all monotonically increasing functions, its real root must be the greatest, thus, the answer is $\boxed{\textbf{(B) }   x^{17}+2018x^{11}+1}$

~isabelchen

Solution 6 (Monotonic Function)

Denote the polynomials in the answer choices by $A(x),B(x),C(x),D(x),$ and $E(x),$ respectively.

Notice that all of the real roots for all answer choices are in $(-1,0)$. The real root of $E(x)$ is $x=-\frac{2018}{2019}\approx-1.000.$, therefore, we can eliminate this polynomial first.

First compare the real roots of $A(x)$ and $C(x)$:

Let the real root of $A(x)$ be $a$, and the real root of $C(x)$ be $b$. \[a^{19}+2018a^{11}+1 = b^{19}+2018b^{13}+1, \quad a^{19}+2018a^{11} = b^{19}+2018b^{13}\] \[\because a^2 < 1, \quad a^{11} < 0, \quad \therefore a^{13}>a^{11}\] \[a^{19}+2018a^{11} < a^{19}+2018a^{13}, \quad b^{19}+2018b^{13} < a^{19}+2018a^{13}\] Let $f(x) = x^{19}+2018x^{13}$. Notice that $f(x)$ is a monotonically increasing function. As $f(b)<f(a)$, $b<a$, $A(x)$ has a bigger real root than $C(x)$.

Similarly, $B(x)$ has a bigger real root than $D(x)$.

To determine the polynomial with the greatest real root, now we only need to compare $A(x)$ and $B(x)$.

Let the real root of $A(x)$ be $a$, and the real root of $B(x)$ be $b$. \[a^{19}+2018a^{11}+1 = b^{17}+2018b^{11}+1, \quad a^{19}+2018a^{11} = b^{17}+2018b^{11}\] \[\because a^2 < 1, \quad a^{17} < 0, \quad \therefore a^{19}>a^{17}\] \[a^{17}+2018a^{11} < a^{19}+2018a^{11}, \quad a^{17}+2018a^{11} < b^{17}+2018b^{11}\] Let $f(x) = x^{17}+2018x^{11}$. Notice that $f(x)$ is a monotonically increasing function. As $f(a)<f(b)$, $a<b$, $B(x)$ has a bigger real root than $A(x)$.

Thus, the answer is $\boxed{\textbf{(B) }  x^{17}+2018x^{11}+1}$

~isabelchen

Solution 7 (Proof by Contradiction)

Denote the polynomials in the answer choices by $A(x),B(x),C(x),D(x),$ and $E(x),$ respectively.

Notice that all of the real roots for all answer choices are in $(-1,0)$. The real root of $E(x)$ is $x=-\frac{2018}{2019}\approx-1.000.$, therefore, we can eliminate this polynomial first.

First compare the real roots of $A(x)$ and $C(x)$:

Let the real root of $A(x)$ be $a$, and the real root of $C(x)$ be $b$. \[a^{19}+2018a^{11}+1 = b^{19}+2018b^{13}+1\] \[a^{11}(a^8+2018) = b^{11}(b^8+2018b^2)\] \[\frac{ a^{11} }{ b^{11} } = \frac{ b^8+2018b^2 }{ a^8+2018 }\] As $b^2<1$, $b^8+2018b^2<b^8+2018$, \[\frac{ a^{11} }{ b^{11} } = \frac{ b^8+2018b^2 }{ a^8+2018 } < \frac{ b^8+2018 }{ a^8+2018 }\]

Assume that $-1<a<b<0$. Thus, \[-1<a^{11}<b^{11}<0, \quad 1<\frac{ a^{11} }{ b^{11} }, \quad 1< \frac{ b^8+2018 }{ a^8+2018 }\]

$\because -1<a<b<0, \quad \therefore 1>a^8>b^8>0, \quad a^8 + 1 > b^8+1, \quad \therefore \frac{b^8+2018}{a^8+2018}<1$

\[\text{ Contradicts with } 1<\frac{b^8+2018}{a^8+2018}\]

$\therefore$ assumption $-1<a<b<0$ is false, $-1<b<a<0$, $A(x)$ has a bigger real root than $C(x)$.

Similarly, $B(x)$ has a bigger real root than $D(x)$.

To determine the polynomial with the greatest real root, now we only need to compare $A(x)$ and $B(x)$.

Let the real root of $A(x)$ be $a$, and the real root of $B(x)$ be $b$. \[a^{19}+2018a^{11}+1 = b^{17}+2018b^{11}+1\] \[a^{19} - b^{17} = 2018( b^{11} - a^{11} )\]

Assume that $-1<b<a<0$. Thus, \[-1<b^{11}<a^{11}<0, \quad b^{11}-a^{11}<0, \quad 2018( b^{11} - a^{11} ) < 0, \quad a^{19} - b^{17} < 0\]

\[-1<b^{17}<a^{17}<0, \quad a^2<1, \quad a^{17}<a^{19}, \quad b^{17}<a^{19}, \quad a^{19} - b^{17} > 0\]

\[\text{ Contradicts with }a^{19} - b^{17} < 0\]

$\therefore$ assumption $-1<b<a<0$ is false, $-1<a<b<0$, $B(x)$ has a bigger real root than $A(x)$.

Thus, the answer is $\boxed{\textbf{(B) }   x^{17}+2018x^{11}+1}$

~isabelchen

Solution 8 (Calculus)

Note that $a(-1)=b(-1)=c(-1)=d(-1) < 0$ and $a(0)=b(0)=c(0)=d(0) > 0$. Calculating the definite integral for each function in the interval $[-1,0]$, we see that $B(x)\rvert^{0}_{-1}$ gives the most negative value. To maximize our real root, we want to maximize the area of the curve under the x-axis, which means we want our integral to be as negative as possible and thus the answer is $\boxed{\textbf{(B) }   x^{17}+2018x^{11}+1}$.

Solution 9 (Calculus)

Newton's Method is used to approximate the zero $x_{1}$ of any real valued function given an estimation for the root $x_{0}: \ x_{1}=x_{0}-{\frac {f(x_{0})}{f'(x_{0})}}.$ After looking at all the options, $x_{0}=-1$ gives a reasonable estimate. For options $\textbf{(A)}$ to $\textbf{(D)},$ we have $f(-1) = -2018$ and the estimation becomes $x_{1}=-1+{\frac {2018}{f'(-1)}}.$ Thus we need to minimize the derivative, giving us $\textbf{(B)}$. Now after comparing $\textbf{(B)}$ and $\textbf{(E)}$ through Newton's method, we see that $\textbf{(B)}$ has the higher root, so the answer is $\boxed{\textbf{(B) }   x^{17}+2018x^{11}+1}$.

~Qcumber

Solution 10 (very fast)

Obviously, eliminate $(E).$ Note that for answer choices $(A)$ and $(B),$ $x \approx \sqrt[11]{\left(-\frac{1}{2018}\right)} > \sqrt[13]{\left(-\frac{1}{2018}\right)},$ with the RHS being approximately the value for $x$ when considering answer choices $(C)$ or $(D),$ so the answer is either $(A)$ or $(B).$ Now, since $x^{19} < x^{17}$ for all negative reals $x$ (we want to choose the option closer to $0$ because then we will have to decrease $x$ less since we estimated), we have $\boxed{(B)}$ as the clear answer.

~ martianrunner

Video Solution by Richard Rusczyk

https://artofproblemsolving.com/videos/amc/2018amc12a/471

~ dolphin7

Video Solution by grogg007 (5 mins, simple solution)

https://www.youtube.com/watch?v=kjjCdK8lPDY

See Also

2018 AMC 12A (ProblemsAnswer KeyResources)
Preceded by
Problem 20
Followed by
Problem 22
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
All AMC 12 Problems and Solutions

These problems are copyrighted © by the Mathematical Association of America, as part of the American Mathematics Competitions. AMC Logo.png