Looking ahead

Algebra Level 3

True or False?

There exists a real matrix A = [ a b c d ] A=\begin{bmatrix} a&b\\c&d\end{bmatrix} such that A 2019 = I 2 A^{2019}=I_2 and a = 2019 a=2019 .

True False

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.

2 solutions

Otto Bretscher
Nov 18, 2018

As David observed, it suffices to find a matrix A A with A 3 = I 2 A^3=I_2 , since A 2019 = ( A 3 ) 673 A^{2019}=(A^3)^{673} . Since A 3 I 2 = ( A I 2 ) ( A 2 + A + I 2 ) A^3-I_2=(A-I_2)(A^2+A+I_2) , it suffices to find a matrix with A 2 + A + I 2 = 0 A^2+A+I_2=0 . By Hamilton-Cayley, it suffices to find a matrix with t r ( A ) = 1 tr(A)=-1 and d e t ( A ) = 1 det(A)=1 (and a = 2019 a=2019 , of course). Such a matrix can be constructed by inspection, for example, A = [ 2019 2019 × 2020 + 1 1 2020 ] A=\begin{bmatrix} 2019& 2019\times 2020 +1\\-1 & -2020 \end{bmatrix} .

(This proof can be presented more elegantly in terms of complex eigenvalues, but I attempted to keep things as basic as possible.)

nice problem. i made a random guess. the hint is the minimal polynomial divides the horrible polynomial x^2019-1. all the roots of the horrible polynomial are distinct so the minimal polynomial has degree 2. rest is construction of the matrix which satisfies the condition.

Srikanth Tupurani - 2 years, 4 months ago
David Vreken
Nov 18, 2018

Cubing the rotation matrix R = [ cos θ sin θ sin θ cos θ ] R = \begin{bmatrix} \cos \theta & -\sin \theta \\ \sin \theta & \cos \theta \end{bmatrix} for θ = 2 π 3 \theta = \frac{2 \pi}{3} will give the identity matrix I 2 I_2 . In other words, B 3 = I 2 B^3 = I_2 for B = [ 1 2 3 2 3 2 1 2 ] B = \begin{bmatrix} -\frac{1}{2} & -\frac{\sqrt{3}}{2} \\ \frac{\sqrt{3}}{2} & -\frac{1}{2} \end{bmatrix} .

This also means that ( C B C 1 ) 3 = I 2 (CBC^{-1})^3 = I_2 for any invertible matrix C = [ a b c d ] C = \begin{bmatrix} a' & b' \\ c' & d' \end{bmatrix} .

Now, if A = [ a b c d ] = C B C 1 A = \begin{bmatrix} a & b \\ c & d \end{bmatrix} = CBC^{-1} , then A 3 = I 2 {A}^3 = I_2 , which in turn means A 2019 = ( A 3 ) 673 = ( I 2 ) 673 = I 2 {A}^{2019} = ({A}^{3})^{673} = (I_2)^{673} = I_2 . All we need now is to find values for a a' , b b' , c c' , and d d' such that a = 2019 a = 2019 , and the statement will be proved true.

Using C 1 = [ d a d b c b a d b c c a d b c a a d b c ] C^{-1} = \begin{bmatrix} \frac{d'}{a'd' - b'c'} & \frac{-b'}{a'd' - b'c'} \\ \frac{-c'}{a'd' - b'c'} & \frac{a'}{a'd' - b'c'} \end{bmatrix} and a little bit of matrix multiplication, we find that a = 1 a d b c ( 1 2 a d + 3 2 b d + 3 2 a c + 1 2 b c ) a = \frac{1}{a'd' - b'c'}(-\frac{1}{2}a'd' + \frac{\sqrt{3}}{2}b'd' + \frac{\sqrt{3}}{2}a'c' + \frac{1}{2}b'c') , b = 1 a d b c ( 1 2 a b 3 2 b 2 3 2 a 2 1 2 a b ) b = \frac{1}{a'd' - b'c'}(\frac{1}{2}a'b' - \frac{\sqrt{3}}{2}b'^2 - \frac{\sqrt{3}}{2}a'^2 - \frac{1}{2}a'b') , c = 1 a d b c ( 1 2 c d + 3 2 b 2 + 3 2 c 2 + 1 2 c d ) c = \frac{1}{a'd' - b'c'}(-\frac{1}{2}c'd' + \frac{\sqrt{3}}{2}b'^2 + \frac{\sqrt{3}}{2}c'^2 + \frac{1}{2}c'd') , and d = 1 a d b c ( 1 2 b c 3 2 b d 3 2 a c 1 2 a d ) d = \frac{1}{a'd' - b'c'}(\frac{1}{2}b'c' - \frac{\sqrt{3}}{2}b'd' - \frac{\sqrt{3}}{2}a'c' - \frac{1}{2}a'd') .

There are several real solutions to obtain a = 2019 a = 2019 , one being a = 1 a' = 1 , b = 4039 3 3 b' = \frac{4039\sqrt{3}}{3} , c = 0 c' = 0 , and d = 1 d' = 1 . These values give us b = 8156762 3 3 b = \frac{8156762 \sqrt{3}}{3} , c = 3 2 c = \frac{\sqrt{3}}{2} , and d = 2020 d = -2020 .

Therefore, [ 2019 8156762 3 3 3 2 2020 ] 2019 = I 2 \begin{bmatrix} 2019 & \frac{8156762 \sqrt{3}}{3} \\ \frac{\sqrt{3}}{2} & -2020 \end{bmatrix}^{2019} = I_2 , which is sufficient to show that the statement is true .

Looks good! A very systematic solution! Thank you!

For the sake of variety, I will post my solution as well.

Otto Bretscher - 2 years, 6 months ago

Log in to reply

Thanks! I would love to see your solution as well.

David Vreken - 2 years, 6 months ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...