Finding a basis for an eigenspace is a fundamental skill in linear algebra, bridging the gap between abstract matrix theory and practical applications in differential equations, quantum mechanics, and data science. Mathematically, it is the null space (or kernel) of the matrix $(A - \lambda I)$. The eigenspace associated with a specific eigenvalue $\lambda$ consists of all eigenvectors corresponding to $\lambda$, along with the zero vector. To find a basis for this space, you must solve a homogeneous linear system and identify the linearly independent vectors that span the solution set Small thing, real impact..
Understanding the Core Concept
Before diving into the computational steps, it is crucial to visualize what you are actually calculating. But an eigenvector $\mathbf{v}$ of a square matrix $A$ satisfies the equation $A\mathbf{v} = \lambda\mathbf{v}$. Rearranging this gives $(A - \lambda I)\mathbf{v} = \mathbf{0}$. In real terms, for a non-trivial solution (where $\mathbf{v} \neq \mathbf{0}$) to exist, the matrix $(A - \lambda I)$ must be singular, meaning its determinant is zero. This condition yields the characteristic polynomial, whose roots are the eigenvalues.
Once an eigenvalue $\lambda$ is confirmed, the set of all vectors $\mathbf{v}$ satisfying $(A - \lambda I)\mathbf{v} = \mathbf{0}$ forms the eigenspace $E_\lambda$. Because this is a solution set to a homogeneous system, it is a valid subspace of $\mathbb{R}^n$ (or $\mathbb{C}^n$). A basis for this subspace is simply a minimal set of linearly independent vectors that span $E_\lambda$. The number of vectors in this basis is the geometric multiplicity of the eigenvalue It's one of those things that adds up..
Step-by-Step Procedure
The process follows a strict algorithmic flow. Mastering each step ensures you avoid common algebraic pitfalls And that's really what it comes down to..
1. Verify the Eigenvalue
If the problem provides the eigenvalue $\lambda$, you can skip the characteristic polynomial calculation. Still, it is best practice to verify that $\det(A - \lambda I) = 0$. If the determinant is non-zero, $\lambda$ is not an eigenvalue, and the eigenspace is trivial (containing only the zero vector), which technically has an empty basis and dimension zero.
2. Construct the Matrix $(A - \lambda I)$
Subtract $\lambda$ from each diagonal entry of matrix $A$. This shifts the spectrum of the matrix so that the eigenvalue of interest maps to zero. $ B = A - \lambda I $ Pay close attention to signs. If $\lambda$ is negative, subtracting a negative number adds to the diagonal entries Easy to understand, harder to ignore..
3. Row Reduce to Reduced Row Echelon Form (RREF)
Perform Gaussian elimination (Gauss-Jordan elimination) on matrix $B$ to obtain its Reduced Row Echelon Form (RREF). While Row Echelon Form (REF) is sufficient to identify pivot columns, RREF makes reading the parametric vector form significantly easier and less prone to error.
- Identify pivot columns (columns containing leading 1s).
- Identify free columns (columns without leading 1s). These correspond to free variables.
4. Write the General Solution in Parametric Vector Form
Express the basic variables (pivot variables) in terms of the free variables. Write the solution vector $\mathbf{v}$ as a linear combination of vectors, where the coefficients are the free variables.
If you have free variables $t_1, t_2, \dots, t_k$, the solution looks like: $ \mathbf{v} = t_1 \mathbf{u}_1 + t_2 \mathbf{u}_2 + \dots + t_k \mathbf{u}_k $
5. Extract the Basis Vectors
The vectors $\mathbf{u}_1, \mathbf{u}_2, \dots, \mathbf{u}_k$ attached to the free parameters form a basis for the eigenspace. These vectors are guaranteed to be linearly independent because each free variable controls a distinct coordinate position (specifically, the coordinate corresponding to its own free column) that no other basis vector controls.
The set ${ \mathbf{u}_1, \mathbf{u}_2, \dots, \mathbf{u}_k }$ is your answer. The dimension of the eigenspace (geometric multiplicity) is $k$.
Worked Example: A 3x3 Matrix
Let’s apply this workflow to a concrete example. Consider the matrix: $ A = \begin{bmatrix} 2 & 1 & 1 \ 0 & 1 & 0 \ 0 & 0 & 1 \end{bmatrix} $
Step 1: Find Eigenvalues The characteristic polynomial is $\det(A - \lambda I)$. $ \det \begin{bmatrix} 2-\lambda & 1 & 1 \ 0 & 1-\lambda & 0 \ 0 & 0 & 1-\lambda \end{bmatrix} = (2-\lambda)(1-\lambda)^2 $ Eigenvalues are $\lambda_1 = 2$ (algebraic multiplicity 1) and $\lambda_2 = 1$ (algebraic multiplicity 2).
Step 2: Basis for $\lambda = 2$ Construct $A - 2I$: $ \begin{bmatrix} 0 & 1 & 1 \ 0 & -1 & 0 \ 0 & 0 & -1 \end{bmatrix} $ Row reduce to RREF:
- Swap Row 1 and Row 2 (or scale Row 2 by -1): $\begin{bmatrix} 0 & 1 & 0 \ 0 & 1 & 1 \ 0 & 0 & -1 \end{bmatrix}$
- $R_2 \leftarrow R_2 - R_1$: $\begin{bmatrix} 0 & 1 & 0 \ 0 & 0 & 1 \ 0 & 0 & -1 \end{bmatrix}$
- $R_3 \leftarrow R_3 + R_2$: $\begin{bmatrix} 0 & 1 & 0 \ 0 & 0 & 1 \ 0 & 0 & 0 \end{bmatrix}$
System: $v_2 = 0$ $v_3 = 0$ $v_1$ is free (let $v_1 = t$).
Parametric Form: $ \mathbf{v} = \begin{bmatrix} t \ 0 \ 0 \end{bmatrix} = t \begin{bmatrix} 1 \ 0 \ 0 \end{bmatrix} $
Basis for $E_2$: $\left{ \begin{bmatrix} 1 \ 0 \ 0 \end{bmatrix} \right}$. Dimension = 1.
Step 3: Basis for $\lambda = 1$ Construct $A - 1I$: $ \begin{bmatrix} 1 & 1 & 1 \ 0 & 0 & 0 \ 0 & 0 & 0 \end{bmatrix} $ This is already in RREF.
System: $v_1 + v_2 + v_3 = 0 \implies v_1 = -v_2 - v_3$. $v_2$ is free (let $v_2 = s$). $v_3$ is free (let $v_3 = t$) Still holds up..
Parametric Form: $ \mathbf{v} = \begin{bmatrix} -s - t \ s \ t \end{bmatrix} = s \begin{bmatrix} -1 \ 1 \ 0 \end{bmatrix} + t \begin{bmatrix} -1 \ 0 \ 1 \end{bmatrix} $
Basis for $E_1$: $\left{ \begin{bmatrix}
