Finding eigenvectors of a simple 2x2 matrix

Click For Summary

Homework Help Overview

The discussion revolves around finding eigenvectors for a 2x2 matrix, specifically the matrix |1 2| |3 4|. The original poster has calculated the eigenvalues but is struggling to derive the corresponding eigenvectors.

Discussion Character

  • Exploratory, Assumption checking, Mathematical reasoning

Approaches and Questions Raised

  • The original poster attempts to substitute the eigenvalues back into the matrix to form equations for the eigenvector but encounters difficulties in solving them. Some participants suggest using a parameter to express the solution, while others question the approach of eliminating variables.

Discussion Status

Participants are actively engaging with the problem, offering guidance on how to approach the equations. There is recognition that the equations may be dependent, leading to an infinite number of solutions, but no consensus on the specific method to resolve the original poster's confusion.

Contextual Notes

The original poster expresses concern about obtaining trivial solutions and is unsure about the implications of the equations derived from the eigenvalue substitution.

The thinker
Messages
54
Reaction score
0

Homework Statement



The matrix is:
|1 2|
|3 4|

The Attempt at a Solution



I've worked out the eigenvalues to be [tex]\stackrel{\underline{5\pm\sqrt{33}}}{2}[/tex]

But when I plug the first eigenvalue back in I get:

|1 - [tex]\stackrel{\underline{5+\sqrt{33}}}{2}[/tex]......2 |
|3......4 - [tex]\stackrel{\underline{5+\sqrt{33}}}{2}[/tex] |

Multiplied by (x,y) = (0,0)

[Sorry I couldn't fit that on the same line as the matrix.]Which in turn gives two equations:

x*(1-[tex]\stackrel{\underline{5+\sqrt{33}}}{2}[/tex]) + 2y = 0

and3x + y*(4-[tex]\stackrel{\underline{5+\sqrt{33}}}{2}[/tex]) = 0Which I can't work out how to solve for the eigenvector.

Is the only solution x=y=0?
 
Physics news on Phys.org
You solve them like you solve any two linear equations. Remember if (x,y) is an eigenvector then a*(x,y) is also an eigenvector. So you are going to have an undetermined parameter which you may as well may make x. Put say, x=1 and solve for y. You should get the same value of y from both equations, since one equation is really a multiple of the other.
 
Thanks for the reply!

Thats exactly what I've been trying to do but X and/or Y always end up canceling and I just get some useless number.
 
The thinker said:
Thanks for the reply!

Thats exactly what I've been trying to do but X and/or Y always end up canceling and I just get some useless number.

Are you trying to eliminate a variable from the two equations and solve for the other? That won't work. If you eliminate x then the resulting equation is 0*y=0. Is that your problem? Like I said before, that's happening because the two equations are really the same. They have an infinite number of solutions. Put x equal to any constant (like x=1) and then solve for y.
 
AH!

Thank you!
 

Similar threads

  • · Replies 12 ·
Replies
12
Views
3K
  • · Replies 19 ·
Replies
19
Views
4K
Replies
2
Views
2K
  • · Replies 7 ·
Replies
7
Views
2K
  • · Replies 2 ·
Replies
2
Views
1K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
Replies
5
Views
2K
  • · Replies 10 ·
Replies
10
Views
2K
Replies
8
Views
3K