Polynomial Problem

Algebra Level 5

If p p and q q are integers such that

x 2 x 1 x^{2}-x-1 is a factor of p x 17 + q x 16 + 1 px^{17}+qx^{16}+1 ,

find p p .


The answer is 987.

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

Lawrence Bush
Nov 25, 2014

Since x 2 x 1 x^{2}-x-1 is a factor of the first polynomial p x 17 + q x 16 + 1 px^{17}+qx^{16}+1 it means that the roots of the first equation are also the roots of the second polynomial.We can solve the first equation by quadratic formula to find :

x 1 = ϕ x_{1}=\phi and x 2 = 1 ϕ x_{2}=1-\phi . where ϕ \phi is the Golden Ratio.

But since our second polynomial contains high degrees it would be better to transform the second root into 1 ϕ \frac{-1}{\phi} .Therefore,

x 1 = ϕ x_{1}=\phi and x 2 = 1 ϕ x_{2}=\frac{-1}{\phi} .

Now,after substituting x x we get the two equations:

p ϕ 17 + q ϕ 16 = 1 p\phi^{17}+q\phi^{16}=-1

q ϕ 16 p ϕ 17 = 1 \frac{q}{\phi^{16}}-\frac{p}{\phi^{17}}=-1

Multiplying the second equation by ϕ 32 \phi^{32} then from the first unchanged equation substract the second one to get:

p ( ϕ 17 + ϕ 15 ) = ϕ 32 1 p(\phi^{17}+\phi^{15})=\phi^{32}-1 . And finally:

p = ϕ 32 1 ϕ 17 + ϕ 15 \boxed{p=\frac{\phi^{32}-1}{\phi^{17}+\phi^{15}}} .

Substitute ϕ = 1 + 5 2 \phi=\frac{1+\sqrt{5}}{2} to get p = 987 p=\boxed{987} .One can also solve this ,i think using Fibonacci numbers because the answer is equal to to F 16 F_{16} where F n F_{n} denotes the n n -th fibonacci number.

Eleganttt..

Ayush Verma - 6 years, 6 months ago
Hugh Sir
Nov 26, 2014

The roots of x 2 x 1 = 0 x^{2}-x-1=0 are α = 1 + 5 2 \alpha=\frac{1+\sqrt{5}}{2} and β = 1 5 2 \beta=\frac{1-\sqrt{5}}{2} .

These roots are also the roots of p x 17 + q x 16 + 1 = 0 px^{17}+qx^{16}+1=0 .

So p α 17 + q α 16 = 1 _ _ _ _ _ _ ( 1 ) p\alpha^{17}+q\alpha^{16} = -1 \space\space\_\_\_\_\_\_(1) ,

and p β 17 + q β 16 = 1 _ _ _ _ _ _ ( 2 ) p\beta^{17}+q\beta^{16} = -1 \space\space\_\_\_\_\_\_(2) .

By multiplying ( 1 ) (1) by β 16 \beta^{16} and using the fact α β = 1 \alpha\beta = -1 ,

p α + q = β 16 p\alpha+q = -\beta^{16} .

By multiplying ( 2 ) (2) by α 16 \alpha^{16} and using the fact α β = 1 \alpha\beta = -1 ,

p β + q = α 16 p\beta+q = -\alpha^{16} .

By subtracting,

p ( α β ) = α 16 β 16 p(\alpha-\beta) = \alpha^{16}-\beta^{16} .

So p = α 16 β 16 α β p = \frac{\alpha^{16}-\beta^{16}}{\alpha-\beta} .

Note that this is the Binet's formula for the 16th Fibonacci number.

Finding the 16th Fibonacci number: 1 , 1 , 2 , 3 , 5 , 8 , 13 , 21 , 34 , 55 , 89 , 144 , 233 , 377 , 610 , 987 , . . . 1,1,2,3,5,8,13,21,34,55,89,144,233,377,610,987,... .

Therefore, p = F 16 = 987 p = F_{16} = 987 .

Alternative Solution:

p = α 16 β 16 α β = ( α 8 + β 8 ) ( α 4 + β 4 ) ( α 2 + β 2 ) ( α + β ) p = \frac{\alpha^{16}-\beta^{16}}{\alpha-\beta} = (\alpha^{8}+\beta^{8})(\alpha^{4}+\beta^{4})(\alpha^{2}+\beta^{2})(\alpha+\beta) .

Since

α + β = 1 \alpha+\beta = 1 ,

α 2 + β 2 = ( α + β ) 2 2 α β = 1 + 2 = 3 \alpha^{2}+\beta^{2} = (\alpha+\beta)^{2}-2\alpha\beta = 1+2 = 3 ,

α 4 + β 4 = ( α 2 + β 2 ) 2 2 α 2 β 2 = 9 2 = 7 \alpha^{4}+\beta^{4} = (\alpha^{2}+\beta^{2})^{2}-2\alpha^{2}\beta^{2} = 9-2 = 7 ,

α 8 + β 8 = ( α 4 + β 4 ) 2 2 α 4 β 4 = 49 2 = 47 \alpha^{8}+\beta^{8} = (\alpha^{4}+\beta^{4})^{2}-2\alpha^{4}\beta^{4} = 49-2 = 47 ,

it follows that p = 47 × 7 × 3 × 1 = 987 p = 47 \times 7 \times 3 \times 1 = 987 .

Nice solution!

Rishu Jaar - 3 years, 7 months ago
Chew-Seong Cheong
Nov 26, 2014

Let f ( x ) = x 2 x 1 f(x) = x^2-x-1 , g ( x ) = p x 17 + q x 16 + 1 g(x) = px^{17}+qx^{16}+1 and the roots of f ( x ) f(x) be φ \varphi . If f ( x ) f(x) is a factor of g ( x ) g(x) , then φ \varphi are also roots of g ( x ) g(x) or g ( φ ) = 0 g(\varphi) = 0 .

f ( φ ) = φ 2 φ 1 = 0 φ 2 = φ + 1 f(\varphi) = \varphi^2 - \varphi\ -1 = 0\quad \Rightarrow \varphi^2 = \varphi +1

φ 4 = ( φ + 1 ) 2 = φ 2 + 2 φ + 1 = 3 φ + 2 \Rightarrow \varphi^4 = (\varphi +1)^2 = \varphi^2 + 2\varphi +1 = 3\varphi +2

φ 8 = ( 3 φ + 2 ) 2 = 9 φ 2 + 12 φ + 4 = 21 φ + 13 \Rightarrow \varphi^8 = (3\varphi +2)^2 = 9 \varphi^2 + 12\varphi +4 = 21 \varphi + 13

φ 16 = ( 21 φ + 13 ) 2 = 441 φ 2 + 546 φ + 169 = 987 φ + 610 \Rightarrow \varphi^{16} = (21 \varphi + 13)^2 = 441 \varphi^2 + 546 \varphi + 169 = 987 \varphi + 610

φ 17 = φ ( 987 φ + 610 ) = 987 ( φ + 1 ) + 610 φ = 1597 φ + 987 \Rightarrow\varphi^{17} = \varphi (987 \varphi + 610) = 987 (\varphi + 1) + 610 \varphi = 1597 \varphi + 987

Therefore, g ( φ ) = p φ 17 + q φ 16 + 1 = p ( 1597 φ + 987 ) + q ( 987 φ + 610 ) + 1 g(\varphi) = p\varphi^{17}+q\varphi^{16}+1 = p(1597 \varphi + 987) + q(987 \varphi + 610) + 1 = ( 1597 p + 987 q ) φ + 987 p + 610 q + 1 = 0 \quad \quad = (1597p + 987q) \varphi + 987p + 610q + 1 = 0

{ 1597 p + 987 q = 0 987 p + 610 q = 1 { 610 ( 1597 p + 987 q ) = 0 987 ( 987 p + 610 q ) = 987 \Rightarrow \begin {cases} 1597p + 987q = 0 \\ 987p + 610q = -1 \end {cases} \quad \Rightarrow \begin {cases} 610(1597p + 987q) = 0 \\ 987(987p + 610q) = -987 \end {cases}

{ 974170 p + 602070 q = 0 974169 p + 602070 q = 987 p = 987 \Rightarrow \begin {cases} 974170p+ 602070q =0 \\ 974169p+602070q=-987 \end {cases} \quad \Rightarrow p = \boxed {987}

Yes, φ = 1 ± 5 2 \varphi = \frac {1\pm \sqrt{5}} {2} is the golden ratio, and φ n = F ( n ) φ + F ( n 1 ) \varphi^n = F(n)\varphi + F(n-1) , where F ( n ) F(n) is the n t h n^{th} Fibonaci number.

Chew-Seong Cheong .I first got equation 1597 p + 987 q = 0 1597p+987q=0 by using formula f n = ( ϕ n τ n ) 5 { f }_{ n }=\frac { \left( { \phi }^{ n }-{ \tau }^{ n } \right) }{ \sqrt { 5 } } .Now using GCD and Extended Euclidean Algorithm I got more generalized form of 1597 + 987 n 987 × 1597 n = 0 1597+987n-987\times 1597n=0 smallest of which we can take is when n = 1 n=1 and p = 987 p=987 .

shivamani patil - 6 years, 6 months ago
Aneesh Kundu
Dec 3, 2014

g ( x ) = x 2 x 1 = 0 g(x)=x^2-x-1=0 x = 1 ± 5 2 \Rightarrow x=\dfrac{1\pm\sqrt{5}}{2} let φ = 1 + 5 2 , ψ = 1 5 2 \varphi=\dfrac{1+\sqrt{5}}{2}, \psi=\dfrac{1-\sqrt{5}}{2} f ( x ) = p x 17 + q x 16 + 1 f(x)=px^{17}+qx^{16}+1 Since g ( x ) g(x) is factor of f ( x ) f(x) , so φ \varphi and ψ \psi are also the roots of f ( x ) f(x) f ( φ ) = p φ 17 + q φ 16 + 1 = 0 f(\varphi)=p\varphi^{17}+q\varphi^{16}+1=0 f ( ψ ) = p ψ 17 + q ψ 16 + 1 = 0 f(\psi)=p\psi^{17}+q\psi^{16}+1=0 The above equations are linear equations in terms of p p and q q . so we use cross-multiplication to solve them. p φ 16 ψ 16 = 1 φ 17 ψ 16 φ 16 ψ 17 \Rightarrow \dfrac{p}{\varphi^{16}-\psi^{16} }= \dfrac{1}{\varphi^{17}\psi^{16} - \varphi^{16}\psi^{17}} p = φ 16 ψ 16 φ 16 ψ 16 ( φ ψ ) \Rightarrow p=\dfrac{\varphi^{16}-\psi^{16}}{\varphi^{16}\psi^{16}(\varphi-\psi)} We know that ψ 16 φ 16 = ( 1 ) 16 = 1 \psi^{16}\varphi^{16}=(-1)^{16}=1 and also that F n = φ n ψ n φ ψ F_{n}=\dfrac{\varphi^{n}-\psi^{n}}{\varphi-\psi} and F 16 = 987 F_{16}=987 Substituting these values, we get p = 987 \boxed{p=987}

Jeremiah Jocson
Nov 25, 2014

try reverse synthetic division :D

Tried but failed

Narasimha Rao B L - 6 years, 6 months ago

When I had to start writing code to manage it (anything more than 3-4 lines is not acceptable in my opinion) I realized there was a better way to do this.

Sidharth Ghoshal - 6 years, 6 months ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...