Source

Adapted from my own handwritten and hand-typed notes from Year 1 to Year 3 of my EEE degree

An eigenvector of a square matrix is a nonzero vector whose direction is not changed by the matrix, it only gets stretched or flipped. The stretch factor is the eigenvalue.

Definition

For an matrix , a number is an eigenvalue if there is a nonzero vector with

Every such nonzero is an eigenvector for . Any nonzero multiple of an eigenvector is also an eigenvector.

Characteristic Equation

Rearranging gives . A nonzero solution exists only when the matrix is singular:

This polynomial equation of degree is the characteristic equation. Its roots are the eigenvalues.

Method

Steps:

  1. Solve for all eigenvalues
  2. For each , solve by row reduction
  3. The solution set is the null space of , called the eigenspace of . Its basis vectors are the eigenvectors

Worked Example

Find the eigenvalues and eigenvectors of .

Step 1:

so and .

Step 2:

  • : gives , so
  • : gives , so

Check: .

Special Cases

Cases to watch for:

  • Complex eigenvalues: a real matrix can have complex eigenvalues, which come in conjugate pairs. E.g. has , so
  • Repeated eigenvalue: an eigenvalue that is a double root may give two independent eigenvectors (a 2D eigenspace) or only one. Count the free variables in step 2 to find out
  • triangular matrix: the eigenvalues are the diagonal entries
  • is an eigenvalue exactly when is singular

Software: [V, D] = eig(A) in MATLAB returns the eigenvectors in and eigenvalues on the diagonal of .

The main use of eigenvectors in this module is Diagonalisation.