Why Am I Getting Only Two Eigenvalues for This Matrix?

  • Thread starter Thread starter DanielFaraday
  • Start date Start date
  • Tags Tags
    Eigenvalues Matrix
Click For Summary

Homework Help Overview

The discussion revolves around finding the eigenvalues of a given 3x3 matrix. The original poster notes that they are obtaining only two eigenvalues instead of the expected three, leading to questions about the algebraic steps taken during the calculation.

Discussion Character

  • Mathematical reasoning, Assumption checking

Approaches and Questions Raised

  • Participants discuss the implications of canceling a term in the determinant calculation and the potential loss of solutions. There are inquiries about the reasoning behind dividing by a specific term and the proper handling of cases where that term equals zero.

Discussion Status

Participants are actively engaging with the original poster's reasoning, providing clarifications and raising questions about the algebraic manipulations. There is a focus on understanding the consequences of canceling terms in the context of finding eigenvalues, with some guidance offered regarding the need to consider cases separately.

Contextual Notes

There is an emphasis on the importance of correctly handling algebraic expressions in the context of eigenvalue calculations, particularly regarding the determinant of a matrix and the implications of division by zero.

DanielFaraday
Messages
86
Reaction score
0

Homework Statement



Find the eigenvalues of the following matrix:

<br /> \left(<br /> \begin{array}{ccc}<br /> 1 &amp; 0 &amp; -3 \\<br /> 1 &amp; 2 &amp; 1 \\<br /> -3 &amp; 0 &amp; 1<br /> \end{array}<br /> \right)<br />

Homework Equations


The Attempt at a Solution



I think I'm forgetting a basic algebra rule or something. I know there are supposed to be 3 eigenvalues, but I am only getting 2.

<br /> 0=\det (\lambda I-A)=\left|<br /> \begin{array}{ccc}<br /> \lambda -1 &amp; 0 &amp; 3 \\<br /> -1 &amp; \lambda -2 &amp; -1 \\<br /> 3 &amp; 0 &amp; \lambda -1<br /> \end{array}<br /> \right|<br />
<br /> =\frac{(\lambda -1)(\lambda -2)(\lambda -1)-3(3)(\lambda -2)}{(\lambda -2)}<br />
<br /> =(\lambda -1)(\lambda -1)-9<br />
<br /> =(\lambda -1)^2-9<br />
<br /> (\lambda -1)^2=9<br />
<br /> \lambda =\pm 3+1<br />
 
Physics news on Phys.org


Where you cancel the (lambda - 2) term, you lose one solution: lambda = 2.

Remember, you have to factor. By canceling the term (lambda - 2), you're basically saying "lambda cannot under any circumstances equal 2". If you want to do that, you have to consider two separate cases: lambda=2 and lambda!=2.

So the three eigenvalues are 3+1=4, -3+1=-2, and 2.
 


Why did you divide by (lambda - 2) in the first place?
 


Thanks for your help.

So let me get this straight. Whenever I divide by a term to cancel it, I count that term as one of my solutions, right? Does this always work? I didn't quite understand what you meant by testing two separate cases.
 


Pengwuino said:
Why did you divide by (lambda - 2) in the first place?

Because it was much easier to do that than it was to multiply everything out and try to factor a third-degree polynomial.
 


Why did you divide by lambda-2 in the second line of your equations? Do you remember the formulas for a determinant of a 3x3 matrix?
 


Oh, duh. Since there are two terms on the RHS and they are constants then the must both be equal to zero separately. Haha.
 


DanielFaraday said:
Because it was much easier to do that than it was to multiply everything out and try to factor a third-degree polynomial.

You can just factor (x-2) to get an easily solvable equation
 


VeeEight said:
Why did you divide by lambda-2 in the second line of your equations? Do you remember the formulas for a determinant of a 3x3 matrix?

I was just doing two steps at once. Bad form on my part.
 
  • #10


DanielFaraday said:
Thanks for your help.

So let me get this straight. Whenever I divide by a term to cancel it, I count that term as one of my solutions, right? Does this always work? I didn't quite understand what you meant by testing two separate cases.

Yes. Think of it this way, let's say you're trying to find the solutions of x for (x-1)(x-2)(x-3)=0. Obviously your solutions are x=1,2, and 3. If you divide out (x-1), you lose a solution and are left with (x-2)(x-3)=0.
 
  • #11


Pengwuino said:
Yes. Think of it this way, let's say you're trying to find the solutions of x for (x-1)(x-2)(x-3)=0. Obviously your solutions are x=1,2, and 3. If you divide out (x-1), you lose a solution and are left with (x-2)(x-3)=0.

Very good point. Thanks!
 
  • #12


DanielFaraday said:
Very good point. Thanks!

Your welcome. Also, to clarify AU's point, when you divide by (lambda-2), your left with that (lambda-2) in the denominator of the left hand side (the 0). Now for any number other then 2, it's going to be 0 divided by some finite number which is 0. However, with lambda = 2, you get 0/0 which is undefined, not 0.
 
  • #13


Pengwuino said:
Your welcome. Also, to clarify AU's point, when you divide by (lambda-2), your left with that (lambda-2) in the denominator of the left hand side (the 0). Now for any number other then 2, it's going to be 0 divided by some finite number which is 0. However, with lambda = 2, you get 0/0 which is undefined, not 0.

Aha. That makes sense now.
 

Similar threads

Replies
2
Views
2K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 6 ·
Replies
6
Views
1K
Replies
19
Views
2K
Replies
8
Views
2K
  • · Replies 4 ·
Replies
4
Views
3K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 7 ·
Replies
7
Views
4K
Replies
2
Views
2K
  • · Replies 2 ·
Replies
2
Views
1K