Telling fibs

Calculus Level 5

Let the matrix with entries a i j a_{ij} be defined as

( a 11 a 12 a 21 a 22 ) = exp ( 1 1 1 0 ) \begin{pmatrix} a_{11} & a_{12} \\ a_{21} & a_{22} \end{pmatrix} = \exp \begin{pmatrix} 1 & 1 \\ 1 & 0 \end{pmatrix}

a 12 + a 21 = 2 d ( e p e 1 / p ) a_{12}+a_{21} = \frac{2}{\sqrt{d}}(e^{p}-e^{-1/p})

and p = a + b c p = \frac{a+\sqrt{b}}{c} , where a , b , c , a,b,c, and d d are positive integers, find a + b + c + d a+b+c+d .


The answer is 13.

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.

1 solution

Prasun Biswas
May 24, 2015

First of all, we should know beforehand how matrix exponential is defined. If we have a square matrix A A , then, matrix exponential is defined using the Maclaurin expansion of e x e^x , namely, as follows:

exp ( A ) = e A = k = 0 A k k ! \exp(A)=e^A=\sum_{k=0}^\infty\frac{A^k}{k!}


Now, back to our problem, let A : = ( 1 1 1 0 ) A:=\begin{pmatrix}1&1\\1&0\end{pmatrix} . Then, the given equation becomes,

( a 11 a 12 a 21 a 22 ) = e A = k = 0 A k k ! = I 2 + k = 1 A k k ! \begin{pmatrix}a_{11}&a_{12}\\ a_{21}&a_{22}\end{pmatrix}=e^A=\sum_{k=0}^\infty\frac{A^k}{k!}=I_2+\sum_{k=1}^\infty\frac{A^k}{k!}

The next part relies on observation. We first check out the first few powers of A A and then make the following claim:

Claim: A k = ( F k + 1 F k F k F k 1 ) k Z + A^k=\begin{pmatrix}F_{k+1}&F_{k}\\ F_k&F_{k-1}\end{pmatrix}~\forall~k\in\Bbb{Z^+} where F k F_k is the k th k^{\textrm{th}} Fibonacci number with F 0 = 0 , F 1 = 1 F_0=0~,~F_1=1 .

This claim is easily proved by standard induction which (if you need the obvious proof) will be given by me in the comments since I don't want to make the solution any messier than it already is.

Anyway, using this claim, our equation becomes,

( a 11 a 12 a 21 a 22 ) = ( 1 + k = 1 F k + 1 k ! k = 1 F k k ! k = 1 F k k ! 1 + k = 1 F k 1 k ! ) \begin{pmatrix}a_{11}&a_{12}\\ a_{21}&a_{22}\end{pmatrix}=\begin{pmatrix}1+\sum_{k=1}^\infty\frac{F_{k+1}}{k!}&\sum_{k=1}^\infty\frac{F_k}{k!}\\ \sum_{k=1}^\infty\frac{F_k}{k!}&1+\sum_{k=1}^\infty\frac{F_{k-1}}{k!}\end{pmatrix}

We take out our relevant part to be summed, that is,

a 12 + a 21 = 2 k = 1 F k k ! = 2 k = 0 F k k ! a_{12}+a_{21}=2\cdot\sum_{k=1}^\infty\frac{F_k}{k!}=2\cdot\sum_{k=0}^\infty\frac{F_k}{k!}

Now, we use the Binet's Fibonacci formula to rewrite this sum into a familiar form that can be simplified using the Maclaurin series for e x e^x . Rewriting it, we have,

a 12 + a 21 = 2 k = 0 ( ϕ k ( ϕ ) k 5 k ! ) = 2 5 [ ( k = 0 ϕ k k ! ) ( k = 0 ( ϕ 1 ) k k ! ) ] a_{12}+a_{21}=2\cdot\sum_{k=0}^\infty\left(\frac{\phi^k-(-\phi)^{-k}}{\sqrt 5\cdot k!}\right)=\frac{2}{\sqrt 5}\left[\left(\sum_{k=0}^\infty\frac{\phi^k}{k!}\right)-\left(\sum_{k=0}^\infty\frac{(-\phi^{-1})^k}{k!}\right)\right]

Using Maclaurin series of e x e^x , i.e., e x = k = 0 x k k ! e^x=\displaystyle\sum_{k=0}^\infty\frac{x^k}{k!} which converges x \forall~x , we have,

a 12 + a 21 = 2 5 ( e ϕ e 1 / ϕ ) a_{12}+a_{21}=\frac{2}{\sqrt 5}\left(e^\phi-e^{-1/\phi}\right)

where ϕ \phi is the golden ratio and has the value ϕ = 1 + 5 2 \phi=\frac{1+\sqrt 5}{2} .

Comparing values, we have a = 1 , b = 5 , c = 2 , d = 5 a=1,b=5,c=2,d=5 . Add them up and you get the answer, which is 13 \boxed{13} .


Clarifications:

  • I 2 I_2 is the identity matrix of order 2 × 2 2\times 2 .
  • We have, by definition, A 0 = I n A^0=I_n for any square matrix A A of order n × n n\times n .

Inductive proof for the claim:

The base case k = 1 k=1 is true because F 0 = 0 F_0=0 and F 1 = F 2 = 1 F_1=F_2=1 , by definition of Fibonacci numbers and the recurrence we know for Fibonacci numbers.

Inductive Hypothesis: Assume that it's true for some k = p k=p , i.e.,

A p = ( F p + 1 F p F p F p 1 ) A^p=\begin{pmatrix}F_{p+1}&F_p\\ F_p&F_{p-1}\end{pmatrix}

Inductive step:

We use the fact that A x = A y A z A^x=A^y\cdot A^z where x = y + z x=y+z and x , y , z Z 0 + x,y,z\in\Bbb{Z^+_0} for any square matrix A A and also use the recursive definition for F k F_k which is,

F k = F k 1 + F k 2 k Z 2 , F 0 = 1 , F 1 = 1 F_k=F_{k-1}+F_{k-2}~\forall~k\in\Bbb{Z_{\geq 2}}~,~F_0=1,F_1=1

A p + 1 = A p A = ( F p + 1 F p F p F p 1 ) ( 1 1 1 0 ) = ( F p + 1 + F p F p + 1 + 0 F p + F p 1 F p + 0 ) = ( F p + 2 F p + 1 F p + 1 F p ) \begin{aligned}A^{p+1}&=A^p\cdot A\\&=\begin{pmatrix}F_{p+1}&F_p\\ F_p&F_{p-1}\end{pmatrix}\cdot \begin{pmatrix}1&1\\ 1&0\end{pmatrix}\\&=\begin{pmatrix}F_{p+1}+F_p&F_{p+1}+0\\ F_p+F_{p-1}&F_p+0\end{pmatrix}=\begin{pmatrix}F_{p+2}&F_{p+1}\\ F_{p+1}&F_p\end{pmatrix}\end{aligned}

Prasun Biswas - 6 years ago

Log in to reply

Nice! Alternatively, you could try doing it by eigenvalues, eigenvectors, and expressing A = S Λ S 1 A = S \Lambda S^{-1} .

Jake Lai - 6 years ago

Log in to reply

Matrix diagonalization, I guess?

Prasun Biswas - 6 years ago

Yeah, I did it using this method. Easy Peasy!

Kartik Sharma - 5 years, 10 months ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...