Finding Eigenpairs

students, today you will learn how to find eigenpairs—the matching pair of an eigenvalue and an eigenvector. This is a core skill in Linear Algebra because eigenpairs reveal special directions in which a matrix acts like simple stretching or shrinking. 🎯

What you will learn

By the end of this lesson, you should be able to:

explain what an eigenpair is and why it matters,
find eigenvalues and eigenvectors for a matrix,
check whether a vector really is an eigenvector,
connect eigenpairs to the larger idea of eigenvalues and eigenvectors,
use examples to show how eigenpairs appear in real calculations.

Why eigenpairs matter

Imagine a transformation that changes vectors in the plane. Most vectors get turned, stretched, or pushed in a complicated way. But sometimes there are special vectors that keep their direction. They may get longer, shorter, or even flip direction, but they still lie on the same line. Those special vectors are eigenvectors. The number that tells how much they stretch or flip is the eigenvalue. Together, they form an eigenpair.

In notation, if $A$ is a matrix, then a nonzero vector $\mathbf{v}$ is an eigenvector if

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

for some scalar $\lambda$. The pair $(\lambda,\mathbf{v})$ is an eigenpair. The vector must be nonzero because the zero vector would satisfy the equation for every $\lambda$, which would not be useful.

This idea is important in engineering, computer graphics, statistics, and systems of equations. For example, a matrix might describe how a shape is stretched. Eigenvectors show the directions that do not rotate, and eigenvalues show the amount of stretch. 📐

Step 1: Set up the eigenvalue equation

To find eigenpairs for a matrix $A$, start with

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

Then move everything to one side:

$$A\mathbf{v}-\lambda\mathbf{v}=\mathbf{0}$$

Because $\lambda\mathbf{v}$ is the same as $\lambda I\mathbf{v}$, where $I$ is the identity matrix, we get

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

This is the key equation. It says that for a nonzero solution $\mathbf{v}$ to exist, the matrix $A-\lambda I$ must be singular, meaning its determinant is zero:

$$\det\left(A-\lambda I\right)=0$$

This equation is called the characteristic equation. Solving it gives the eigenvalues. After that, each eigenvalue is plugged back into $\left(A-\lambda I\right)\mathbf{v}=\mathbf{0}$ to find the matching eigenvectors.

Step 2: Find the eigenvalues

Let’s use a simple example:

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

We solve

$$\det\left(A-\lambda I\right)=0$$

First form

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

Now compute the determinant:

$$\det\left(A-\lambda I\right)=\left(2-\lambda\right)^2-1$$

Set it equal to zero:

$$\left(2-\lambda\right)^2-1=0$$

Factor:

$$\left(2-\lambda-1\right)\left(2-\lambda+1\right)=0$$

$$\left(1-\lambda\right)\left(3-\lambda\right)=0$$

Therefore the eigenvalues are

$$\lambda=1 \quad \text{and} \quad \lambda=3$$

These are the only values that can produce nonzero eigenvectors. ✅

Step 3: Find the eigenvector for each eigenvalue

Now we solve $\left(A-\lambda I\right)\mathbf{v}=\mathbf{0}$ for each eigenvalue.

For $\lambda=3$

Substitute into the matrix:

$$A-3I=\begin{pmatrix}-1 & 1\\1 & -1\end{pmatrix}$$

Let

$$\mathbf{v}=\begin{pmatrix}x\y\end{pmatrix}$$

Then

$$\begin{pmatrix}-1 & 1\\1 & -1\end{pmatrix}\begin{pmatrix}x\y\end{pmatrix}=\begin{pmatrix}0\\0\end{pmatrix}$$

This gives the equation

$$-x+y=0$$

$$y=x$$

Any nonzero vector of the form

$$\begin{pmatrix}x\x\end{pmatrix}$$

is an eigenvector for $\lambda=3$. A convenient choice is

$$\begin{pmatrix}1\\1\end{pmatrix}$$

So one eigenpair is

$$\left(3,\begin{pmatrix}1\\1\end{pmatrix}\right)$$

For $\lambda=1$

Now substitute $\lambda=1$:

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

Solve

$$\begin{pmatrix}1 & 1\\1 & 1\end{pmatrix}\begin{pmatrix}x\y\end{pmatrix}=\begin{pmatrix}0\\0\end{pmatrix}$$

This gives

$$x+y=0$$

$$y=-x$$

Any nonzero vector of the form

$$\begin{pmatrix}x\\-x\end{pmatrix}$$

is an eigenvector for $\lambda=1$. A simple choice is

$$\begin{pmatrix}1\\-1\end{pmatrix}$$

So another eigenpair is

$$\left(1,\begin{pmatrix}1\\-1\end{pmatrix}\right)$$

How to check an eigenpair

students, once you think you have an eigenpair, it is smart to check it directly by multiplying. This is like verifying an answer in math class.

Check the pair $\left(3,\begin{pmatrix}1\\1\end{pmatrix}\right)$:

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

and

$$3\begin{pmatrix}1\\1\end{pmatrix}=\begin{pmatrix}3\\3\end{pmatrix}$$

They match, so it is correct.

This check works for any candidate eigenpair. If the result of $A\mathbf{v}$ is exactly $\lambda\mathbf{v}$, then the pair is valid. If the directions are different, then the vector is not an eigenvector for that eigenvalue.

Patterns that help you find eigenpairs faster

Some matrices have special patterns that make eigenpairs easier to spot.

Symmetric matrices

The example matrix

$$\begin{pmatrix}2 & 1\\1 & 2\end{pmatrix}$$

is symmetric because it equals its transpose. Symmetric matrices often have nice eigenvalues and eigenvectors, and their eigenvectors can be chosen to be orthogonal. In this case,

$$\begin{pmatrix}1\\1\end{pmatrix} \quad \text{and} \quad \begin{pmatrix}1\\-1\end{pmatrix}$$

are perpendicular because their dot product is

$$1\cdot 1+1\cdot(-1)=0$$

Diagonal matrices

$$A=\begin{pmatrix}a & 0\\0 & d\end{pmatrix}$$

then the standard basis vectors are eigenvectors. For example,

$$\begin{pmatrix}a & 0\\0 & d\end{pmatrix}\begin{pmatrix}1\\0\end{pmatrix}=\begin{pmatrix}a\\0\end{pmatrix}=a\begin{pmatrix}1\\0\end{pmatrix}$$

and

$$\begin{pmatrix}a & 0\\0 & d\end{pmatrix}\begin{pmatrix}0\\1\end{pmatrix}=\begin{pmatrix}0\d\end{pmatrix}=d\begin{pmatrix}0\\1\end{pmatrix}$$

So the eigenpairs are easy to see.

Common mistakes to avoid

A few errors show up often when finding eigenpairs:

forgetting that an eigenvector must be nonzero,
solving for eigenvectors before finding eigenvalues,
making algebra mistakes in the determinant,
mixing up the eigenvalue $\lambda$ with the vector $\mathbf{v}$,
stopping too early and not checking the answer.

A helpful habit is to always verify your final answer by substitution into $A\mathbf{v}=\lambda\mathbf{v}$. 🧠

How eigenpairs fit into the bigger topic

Eigenpairs are the heart of the study of eigenvalues and eigenvectors. The broader topic asks questions like:

Which matrices have eigenvalues?
How many eigenvalues can a matrix have?
What do eigenvectors tell us about the matrix?
How can eigenpairs simplify computations?

Finding eigenpairs is the practical skill that lets you answer those questions. Once you know the eigenpairs, you can understand how a matrix behaves along special directions. In advanced topics, eigenpairs are used in diagonalization, differential equations, Markov chains, and data analysis. That is why this lesson is a foundation for many later ideas.

Conclusion

Finding eigenpairs means finding values $\lambda$ and nonzero vectors $\mathbf{v}$ such that

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

The process is:

write $\left(A-\lambda I\right)\mathbf{v}=\mathbf{0}$,
solve $\det\left(A-\lambda I\right)=0$ for eigenvalues,
substitute each eigenvalue back in,
solve for the matching eigenvectors,
check your work by multiplying.

students, if you remember that eigenvectors are the special directions and eigenvalues are the stretching factors, the whole topic becomes much easier to understand. Finding eigenpairs is not just a formula to memorize—it is a way to see how matrices act in a clean and powerful way. 🌟

Study Notes

An eigenpair is a pair $(\lambda,\mathbf{v})$ where $\mathbf{v}\neq\mathbf{0}$ and $A\mathbf{v}=\lambda\mathbf{v}$.
To find eigenvalues, solve $\det\left(A-\lambda I\right)=0$.
To find eigenvectors, solve $\left(A-\lambda I\right)\mathbf{v}=\mathbf{0}$ for each eigenvalue.
An eigenvector keeps its direction after multiplication by a matrix; only its length may change or flip.
The zero vector is never an eigenvector.
Always check a candidate pair by substituting into $A\mathbf{v}=\lambda\mathbf{v}$.
In the example matrix $\begin{pmatrix}2 & 1\\1 & 2\end{pmatrix}$, the eigenpairs are $\left(3,\begin{pmatrix}1\\1\end{pmatrix}\right)$ and $\left(1,\begin{pmatrix}1\\-1\end{pmatrix}\right)$.
Eigenpairs are central to diagonalization and many applications in science and technology.