finding eigenvalues


This example investigates eigenvaluesMathworldPlanetmathPlanetmathPlanetmathPlanetmath and the similarity transformationMathworldPlanetmath used to diagonalize matrices. We seek the eigenvalues of the matrix A below. Afterward, we can transform this matrix into a diagonal matrixMathworldPlanetmath which has many useful applications.

A=(2112)

Here, we need to solve the corresponding matrix equation;

(2112)⁢(x1x2)=λ⁢(x1x2)

or

A⁢X=λ⁢X

rearranging gives

A⁢X-λ⁢X=0

or

(A-λ⁢I)⁢X=0

We seek the values for λ and X. First, we need to solve the characteristic equationMathworldPlanetmathPlanetmath of A. We do this by finding d⁢e⁢t⁢(A-λ⁢I). First, calculating A-λ⁢I gives;

A-λ⁢I=(2-λ112-λ)

Next, calculating d⁢e⁢t⁢(A-λ⁢I) yields

d⁢e⁢t⁢(A-λ⁢I)=(2-λ)2-1=λ2-4⁢λ+3=(λ-1)⁢(λ-3)=0

Substituting λ=1 into (A-λ⁢I)⁢X gives…

{x1+x2=0x1+x2=0

so that x2=-x1 and the corresponding eigenvectorMathworldPlanetmathPlanetmathPlanetmath is

(t-t)=t⁢(1-1)

where t≠0.
Substituting λ=3 gives…

{-x1+x2=0x1-x2=0

so that x2=x1 and the corresponding eigenvector is

(tt)=t⁢(11)

where t≠0.
Finally, to diagonalize A we let the eigenvectors be the columns of a new matrix

P=(1-111)

and then since our eigenvectors are linearly independentMathworldPlanetmath we can also find;

P-1=12⁢(11-11)

then we create a diagonal matrix as follows…

D=P-1⁢A⁢P=(1003)

Computing powers of A is a very useful application of D. Solving for A lets us compute powers of A

A=P⁢D⁢P-1

so that

An=P⁢Dn⁢P-1

or

An=P⁢(1n003n)⁢P-1

Title finding eigenvalues
Canonical name FindingEigenvalues
Date of creation 2013-03-22 15:52:35
Last modified on 2013-03-22 15:52:35
Owner PrimeFan (13766)
Last modified by PrimeFan (13766)
Numerical id 6
Author PrimeFan (13766)
Entry type Example
Classification msc 15A18