Skip to content

How does a matrix transform the plane?

Determinant, inverse, eigenvalues and products of 2 × 2 and 3 × 3 matrices, with a 2 × 2 matrix drawn acting on the plane.

Size
Operation

Matrix A

Determinant det A

3

A scales every area by 3 and keeps the orientation.

Key values as a table
VectorAfter A
î = (1, 0)(2, 1)
ĵ = (0, 1)(1, 2)
î + ĵ = (1, 1)(3, 3)

Properties

Trace
4
Eigenvalues
3 and 1
Inverse A⁻¹
exists, shown below

Inverse A⁻¹

Inverse A⁻¹
Column 1Column 2
0.666667−0.333333
−0.3333330.666667

Determinant, inverse, eigenvalues and products of 2 × 2 and 3 × 3 matrices, with a 2 × 2 matrix drawn acting on the plane.

How to use

  1. Choose 2 × 2 or 3 × 3, and A alone or A × B.
  2. Type the entries row by row; a₁₂ is row 1, column 2.
  3. Read the determinant (or the product), the trace, the eigenvalues and the inverse. For a 2 × 2 matrix, drag the tips of Aî and Aĵ to change its columns and watch the plane move.

How it works

A matrix is a recipe for moving every point of the plane (or of space) at once. Its columns say where the unit vectors go: A = [a b; c d] sends î = (1, 0) to (a, c) and ĵ = (0, 1) to (b, d), and every other point follows, so grid lines stay straight and evenly spaced. Multiplying A·B means doing B first and then A; each entry of the product is a row of A times a column of B.

The determinant, det A = ad − bc for 2 × 2, is how much A scales areas (volumes for 3 × 3), negative when it flips the orientation like a mirror. When it is 0, A squashes the plane flat and nothing can undo it: there is no inverse. Otherwise A⁻¹ = (1/det A)·[d −b; −c a]. For example [2 1; 1 2] has det 3, and A⁻¹ = [2/3 −1/3; −1/3 2/3].

An eigenvector is a direction A only stretches, by its eigenvalue λ, the solutions of λ² − (trace)λ + det = 0. [2 1; 1 2] stretches (1, 1) by 3 and leaves (1, −1) as it is (λ = 1). A rotation has no such real direction: its eigenvalues are a complex pair. Sources: OpenStax Precalculus 2e, sections 9.5–9.8; Gilbert Strang, MIT OpenCourseWare 18.06 Linear Algebra.

Worked example

What are the determinant, inverse and eigenvalues of [2 1; 1 2]?

  1. Determinant: ad − bc = 2·2 − 1·1 = 3.
  2. Inverse: (1/3)·[2 −1; −1 2] = [0.6667 −0.3333; −0.3333 0.6667].
  3. Eigenvalues: λ² − 4λ + 3 = 0, so λ = 3 and λ = 1.

det = 3, trace = 4, eigenvalues 3 and 1.

Frequently asked questions

Why is A × B not the same as B × A?

Doing B and then A is usually a different motion from A and then B, like rotating and then stretching versus stretching and then rotating. Try both orders: the products differ, but their determinants are equal.

What does a negative determinant mean?

The matrix flips the plane over: what was turning counter-clockwise turns clockwise. The size of the determinant is still the area factor.

Why are eigenvalues only shown for 2 × 2?

A 2 × 2 matrix needs only a quadratic equation, solved exactly here and drawn. For 3 × 3 the trace, determinant and inverse are shown; eigenvalues need a cubic.

How do I solve a system of equations with this?

If Ax = b, then x = A⁻¹b when A has an inverse. The linear system solver does the whole thing for you.

What does the trace of a matrix tell me?

The trace is the sum of the diagonal entries. It equals the sum of the eigenvalues, and together with the determinant it gives both of them for a 2 × 2 matrix.

Send feedback