Linear Algebra

Linear Combination of Eigenvectors is Not an Eigenvector

Problems and Solutions of Eigenvalue, Eigenvector in Linear Algebra

Problem 258

Suppose that $\lambda$ and $\mu$ are two distinct eigenvalues of a square matrix $A$ and let $\mathbf{x}$ and $\mathbf{y}$ be eigenvectors corresponding to $\lambda$ and $\mu$, respectively.
If $a$ and $b$ are nonzero numbers, then prove that $a \mathbf{x}+b\mathbf{y}$ is not an eigenvector of $A$ (corresponding to any eigenvalue of $A$).

Add to solve later

Hint.

We use the following fact in the proof.
Fact: Two eigenvectors corresponding to distinct eigenvalues are linearly independent.

Proof.

Seeking a contradiction, we assume that $a \mathbf{x}+b\mathbf{y}$ is an eigenvector corresponding to an eigenvalue $\zeta$.
Thus we have
\begin{align*}
A(a \mathbf{x}+b\mathbf{y})=\zeta (a \mathbf{x}+b\mathbf{y}). \tag{*}
\end{align*}

We calculate the left hand side of this equality as follows.
We have
\begin{align*}
A(a \mathbf{x}+b\mathbf{y})&=a A\mathbf{x}+bA\mathbf{y}\\
&=a\lambda \mathbf{x}+b\mu \mathbf{y}
\end{align*}
since $A\mathbf{x}=\lambda \mathbf{x}, A\mathbf{y}=\mu \mathbf{y}$ by defintion.

Therefore, from (*) we obtain
\begin{align*}
a(\lambda -\zeta) \mathbf{x}+b(\mu-\zeta)\mathbf{y}=\mathbf{0}.
\end{align*}

Recall that eigenvectors corresponding to distinct eigenvalues are linearly independent. Thus $\mathbf{x}$ and $\mathbf{y}$ are linearly independent.

Thus, the coefficients of the above linear combinations must be zero:
\[a(\lambda -\zeta)=0 \text{ and } b(\mu-\zeta)=0.\]

Since $a\neq0, b\neq 0$, this implies that we have
\[\lambda=\zeta=\mu,\] and this is a contradiction because $\lambda$ and $\mu$ are supposed to be distinct.

Hence, $a \mathbf{x}+b\mathbf{y}$ cannot be an eigenvector of any eigenvalue of $A$.

Add to solve later

More from my site

Two Eigenvectors Corresponding to Distinct Eigenvalues are Linearly Independent Let $A$ be an $n\times n$ matrix. Suppose that $\lambda_1, \lambda_2$ are distinct eigenvalues of the matrix $A$ and let $\mathbf{v}_1, \mathbf{v}_2$ be eigenvectors corresponding to $\lambda_1, \lambda_2$, respectively. Show that the vectors $\mathbf{v}_1, \mathbf{v}_2$ are […]
Two Subspaces Intersecting Trivially, and the Direct Sum of Vector Spaces. Let $V$ and $W$ be subspaces of $\R^n$ such that $V \cap W =\{\mathbf{0}\}$ and $\dim(V)+\dim(W)=n$. (a) If $\mathbf{v}+\mathbf{w}=\mathbf{0}$, where $\mathbf{v}\in V$ and $\mathbf{w}\in W$, then show that $\mathbf{v}=\mathbf{0}$ and $\mathbf{w}=\mathbf{0}$. (b) If $B_1$ is a […]
Compute Determinant of a Matrix Using Linearly Independent Vectors Let $A$ be a $3 \times 3$ matrix. Let $\mathbf{x}, \mathbf{y}, \mathbf{z}$ are linearly independent $3$-dimensional vectors. Suppose that we have \[A\mathbf{x}=\begin{bmatrix} 1 \\ 0 \\ 1 \end{bmatrix}, A\mathbf{y}=\begin{bmatrix} 0 \\ 1 \\ 0 […]
If Vectors are Linearly Dependent, then What Happens When We Add One More Vectors? Suppose that $\mathbf{v}_1, \mathbf{v}_2, \dots, \mathbf{v}_r$ are linearly dependent $n$-dimensional real vectors. For any vector $\mathbf{v}_{r+1} \in \R^n$, determine whether the vectors $\mathbf{v}_1, \mathbf{v}_2, \dots, \mathbf{v}_r, \mathbf{v}_{r+1}$ are linearly […]
Given Eigenvectors and Eigenvalues, Compute a Matrix Product (Stanford University Exam) Suppose that $\begin{bmatrix} 1 \\ 1 \end{bmatrix}$ is an eigenvector of a matrix $A$ corresponding to the eigenvalue $3$ and that $\begin{bmatrix} 2 \\ 1 \end{bmatrix}$ is an eigenvector of $A$ corresponding to the eigenvalue $-2$. Compute $A^2\begin{bmatrix} 4 […]
Given All Eigenvalues and Eigenspaces, Compute a Matrix Product Let $C$ be a $4 \times 4$ matrix with all eigenvalues $\lambda=2, -1$ and eigensapces \[E_2=\Span\left \{\quad \begin{bmatrix} 1 \\ 1 \\ 1 \\ 1 \end{bmatrix} \quad\right \} \text{ and } E_{-1}=\Span\left \{ \quad\begin{bmatrix} 1 \\ 2 \\ 1 \\ 1 […]
$Eigenvalues of Orthogonal Matrices Have Length 1. Every $3\times 3$ Orthogonal Matrix Has 1 as an Eigenvalue$ Eigenvalues of Orthogonal Matrices Have Length 1. Every $3\times 3$ Orthogonal Matrix Has 1 as an Eigenvalue (a) Let $A$ be a real orthogonal $n\times n$ matrix. Prove that the length (magnitude) of each eigenvalue of $A$ is $1$. (b) Let $A$ be a real orthogonal $3\times 3$ matrix and suppose that the determinant of $A$ is $1$. Then prove that $A$ has $1$ as an […]
Eigenvalues of a Hermitian Matrix are Real Numbers Show that eigenvalues of a Hermitian matrix $A$ are real numbers. (The Ohio State University Linear Algebra Exam Problem) We give two proofs. These two proofs are essentially the same. The second proof is a bit simpler and concise compared to the first one. […]