Eigenvalues and Eigenvectors: What They Mean

Most vectors change direction when you multiply them by a matrix. Eigenvectors are the special ones that do not: the matrix only stretches, shrinks or flips them. That simple idea powers Google's original PageRank, vibration analysis, quantum mechanics and principal component analysis. Here is what eigenvalues and eigenvectors really mean and how to compute them step by step.

The Definition

Let \( A \) be an \( n \times n \) matrix. A nonzero vector \( \mathbf{v} \) is an eigenvector of \( A \) if

$$ A\mathbf{v} = \lambda\mathbf{v} $$

for some scalar \( \lambda \). The number \( \lambda \) is the corresponding eigenvalue.

What It Means Geometrically

Think of a matrix as a transformation that moves every vector in the plane. An eigenvector lies on a line that the transformation maps onto itself. The eigenvalue tells you what happens along that line:

• \( \lambda > 1 \): vectors are stretched.
• \( 0 < \lambda < 1 \): vectors are shrunk.
• \( \lambda < 0 \): vectors are flipped to the opposite direction.
• \( \lambda = 0 \): vectors are squashed to zero, so \( A \) is not invertible.

Eigenvectors are the "natural axes" of a transformation. In those directions the matrix behaves like simple multiplication by a number.

How to Find Them

Rewrite \( A\mathbf{v} = \lambda\mathbf{v} \) as

$$ (A - \lambda I)\mathbf{v} = \mathbf{0}. $$

For a nonzero solution \( \mathbf{v} \) to exist, the matrix \( A - \lambda I \) must be singular, so

$$ \det(A - \lambda I) = 0. $$

This is the characteristic equation. The procedure:

1. Form \( A - \lambda I \) by subtracting \( \lambda \) from each diagonal entry.
2. Set its determinant to zero and solve for \( \lambda \).
3. For each \( \lambda \), solve \( (A - \lambda I)\mathbf{v} = \mathbf{0} \) to find the eigenvectors.
4. Check with \( A\mathbf{v} = \lambda\mathbf{v} \).

Worked Example 1: A 2×2 Matrix

Find the eigenvalues and eigenvectors of

$$ A = \begin{pmatrix} 4 & 1 \\ 2 & 3 \end{pmatrix}. $$

Step 1: Form \( A - \lambda I \).

$$ A - \lambda I = \begin{pmatrix} 4 - \lambda & 1 \\ 2 & 3 - \lambda \end{pmatrix}. $$

Step 2: Characteristic equation.

$$ (4 - \lambda)(3 - \lambda) - (1)(2) = \lambda^2 - 7\lambda + 12 - 2 = \lambda^2 - 7\lambda + 10 = 0. $$

Factorising: \( (\lambda - 5)(\lambda - 2) = 0 \), so \( \lambda_1 = 5 \) and \( \lambda_2 = 2 \).

Step 3a: Eigenvector for \( \lambda = 5 \).

$$ A - 5I = \begin{pmatrix} -1 & 1 \\ 2 & -2 \end{pmatrix}. $$

The first row gives \( -x + y = 0 \), so \( y = x \). Taking \( x = 1 \):

$$ \mathbf{v}_1 = \begin{pmatrix} 1 \\ 1 \end{pmatrix}. $$

Step 3b: Eigenvector for \( \lambda = 2 \).

$$ A - 2I = \begin{pmatrix} 2 & 1 \\ 2 & 1 \end{pmatrix}. $$

The first row gives \( 2x + y = 0 \), so \( y = -2x \). Taking \( x = 1 \):

$$ \mathbf{v}_2 = \begin{pmatrix} 1 \\ -2 \end{pmatrix}. $$

Step 4: Check.

$$ A\mathbf{v}_1 = \begin{pmatrix} 4 + 1 \\ 2 + 3 \end{pmatrix} = \begin{pmatrix} 5 \\ 5 \end{pmatrix} = 5\mathbf{v}_1, \qquad A\mathbf{v}_2 = \begin{pmatrix} 4 - 2 \\ 2 - 6 \end{pmatrix} = \begin{pmatrix} 2 \\ -4 \end{pmatrix} = 2\mathbf{v}_2. $$

Both are correct. Any nonzero multiple, such as \( (3, 3) \), is also an eigenvector for \( \lambda = 5 \).

Two Quick Checks

For any square matrix:

• The sum of the eigenvalues equals the trace (sum of diagonal entries). Here \( 5 + 2 = 7 = 4 + 3 \).
• The product of the eigenvalues equals the determinant. Here \( 5 \times 2 = 10 = 12 - 2 \).

Use these to catch errors in seconds.

Worked Example 2: A Symmetric Matrix

Find the eigenvalues and eigenvectors of

$$ B = \begin{pmatrix} 2 & 1 \\ 1 & 2 \end{pmatrix}. $$

Step 1: Characteristic equation.

$$ (2 - \lambda)^2 - 1 = 0 \quad\Longrightarrow\quad 2 - \lambda = \pm 1 \quad\Longrightarrow\quad \lambda = 3 \ \text{or} \ \lambda = 1. $$

Check: trace \( = 4 = 3 + 1 \) and determinant \( = 3 = 3 \times 1 \).

Step 2: \( \lambda = 3 \). \( B - 3I = \begin{pmatrix} -1 & 1 \\ 1 & -1 \end{pmatrix} \), giving \( y = x \), so \( \mathbf{v}_1 = (1, 1) \).

Step 3: \( \lambda = 1 \). \( B - I = \begin{pmatrix} 1 & 1 \\ 1 & 1 \end{pmatrix} \), giving \( y = -x \), so \( \mathbf{v}_2 = (1, -1) \).

Observation: \( \mathbf{v}_1 \cdot \mathbf{v}_2 = 1 - 1 = 0 \), so the eigenvectors are perpendicular. This is always true for symmetric matrices with distinct eigenvalues. Geometrically, \( B \) stretches by 3 along the line \( y = x \) and leaves the line \( y = -x \) unchanged.

Why Eigenvalues Matter

• Powers of matrices: if \( A\mathbf{v} = \lambda\mathbf{v} \), then \( A^k\mathbf{v} = \lambda^k\mathbf{v} \). For example, \( A^{3}\mathbf{v}_1 = 5^3\mathbf{v}_1 = (125, 125) \) without any matrix multiplication.
• Diagonalisation: with enough independent eigenvectors, \( A = PDP^{-1} \), where \( D \) holds the eigenvalues.
• Differential equations: systems \( \mathbf{x}' = A\mathbf{x} \) have solutions built from \( e^{\lambda t}\mathbf{v} \).
• Stability and data science: eigenvalues decide whether systems grow or decay, and PCA uses eigenvectors of a covariance matrix to find the most important directions in data.

Common Mistakes

• Accepting the zero vector as an eigenvector. Eigenvectors must be nonzero.
• Subtracting \( \lambda \) from every entry instead of only the diagonal.
• Arithmetic slips in the determinant. Always confirm with the trace and determinant checks.
• Thinking eigenvectors are unique. Any nonzero scalar multiple works.

Key Takeaways

• An eigenvector keeps its direction under \( A \); the eigenvalue is its scaling factor.
• Solve \( \det(A - \lambda I) = 0 \) for eigenvalues, then \( (A - \lambda I)\mathbf{v} = \mathbf{0} \) for eigenvectors.
• The eigenvalues sum to the trace and multiply to the determinant.
• Symmetric matrices have real eigenvalues and perpendicular eigenvectors.

If eigenvalue problems or diagonalisation are giving you trouble, Amathify's experts and PhD scholars can help. Send your question on WhatsApp or through the "Ask a Question" form for a clear, step-by-step solution.