Lesson 2 of 8 · 15 min

Inverses and matrix equations

Find and check inverses of 3×33 \times 3 and larger matrices with technology, and solve AX=B\bm{AX} = \bm{B}, XA=B\bm{XA} = \bm{B} and AX+BX=C\bm{AX} + \bm{BX} = \bm{C} by multiplying by an inverse on the correct side.

What you will be able to do

  • Find the inverse of a 3×33 \times 3 or larger matrix with technology, and check it by multiplying.
  • Find an inverse from an identity such as AB=kI\bm{AB} = k\bm{I}.
  • Solve matrix equations of the forms AX=B\bm{AX} = \bm{B}, XA=B\bm{XA} = \bm{B} and AX+BX=C\bm{AX} + \bm{BX} = \bm{C}.

The inverse of a square matrix A\bm{A} is the matrix A−1\bm{A}^{-1} that undoes it:

AA−1=I=A−1A\bm{AA}^{-1} = \bm{I} = \bm{A}^{-1}\bm{A}

For 2×22 \times 2 matrices you had a formula. For bigger matrices there is one too, but it takes nine 2×22 \times 2 determinants for a 3×33 \times 3 matrix, so the syllabus expects you to use technology. What you must know is when an inverse exists, how to check one, and how to use it.

Key idea

When does the inverse exist?

A square matrix of any order has an inverse exactly when det⁡(A)≠0\det(\bm{A}) \neq 0. A singular matrix has no inverse, and the calculator reports an error such as "singular matrix".

To check that B\bm{B} is the inverse of A\bm{A}, show that AB=I\bm{AB} = \bm{I}. (For square matrices, AB=I\bm{AB} = \bm{I} forces BA=I\bm{BA} = \bm{I} too.)

Interactive

The inverse of a 3 × 3 matrix

This matrix has determinant 1212, so its inverse has twelfths in it. Change an entry to make the determinant 00: what happens to the inverse?

Worked example

Find and check a 3 × 3 inverse with technology

  • Simple familiar
  • Technology active

Let A=[20113−1021]\bm{A} = \begin{bmatrix} 2 & 0 & 1 \\ 1 & 3 & -1 \\ 0 & 2 & 1 \end{bmatrix}. Show that A\bm{A} is non-singular, and use technology to find A−1\bm{A}^{-1} exactly.

3 steps

Try it yourself first, then check the working one step at a time.

Method

  1. State det⁡(A)≠0\det(\bm{A}) \neq 0 first.

  2. Calculate A−1\bm{A}^{-1} with technology and convert to exact fractions.

  3. Check with AA−1=I\bm{AA}^{-1} = \bm{I}.

Watch out for

  • Copying rounded decimals such as 0.41670.4167 instead of 512\frac{5}{12} when an exact answer is asked for.

How the calculator finds a 3 × 3 inverse

Replace every element of A\bm{A} by its cofactor to get the cofactor matrix C\bm{C}. Transpose it (swap rows and columns) to get the adjugate, adj⁡(A)=CT\operatorname{adj}(\bm{A}) = \bm{C}^{T}. Then

A−1=1det⁡(A)adj⁡(A)\bm{A}^{-1} = \frac{1}{\det(\bm{A})}\operatorname{adj}(\bm{A})

For a 2×22 \times 2 matrix the adjugate is [d−b−ca]\begin{bmatrix} d & -b \\ -c & a \end{bmatrix}, which is the Unit 1 formula. Expanding Aadj⁡(A)\bm{A}\operatorname{adj}(\bm{A}) gives det⁡(A)\det(\bm{A}) on the diagonal and 00 elsewhere, which is why the formula works. You won't be asked to do this by hand for a 3×33 \times 3 matrix, but it explains why the inverse needs det⁡(A)≠0\det(\bm{A}) \neq 0.

Square matrices A\bm{A} and B\bm{B} satisfy AB=3I\bm{AB} = 3\bm{I}. Which of these is A−1\bm{A}^{-1}?

Choose one answer

Matrix equations with bigger matrices work exactly as in Unit 1. There is no matrix division, so you multiply by an inverse, and because AB≠BA\bm{AB} \neq \bm{BA} in general, the side matters.

Solving matrix equations (each needs the inverted matrix to be non-singular)
EquationDo this to both sidesSolution
AX=B\bm{AX} = \bm{B}Pre-multiply by A−1\bm{A}^{-1}X=A−1B\bm{X} = \bm{A}^{-1}\bm{B}
XA=B\bm{XA} = \bm{B}Post-multiply by A−1\bm{A}^{-1}X=BA−1\bm{X} = \bm{BA}^{-1}
AX+BX=C\bm{AX} + \bm{BX} = \bm{C}Factor: (A+B)X=C(\bm{A} + \bm{B})\bm{X} = \bm{C}, then pre-multiplyX=(A+B)−1C\bm{X} = (\bm{A} + \bm{B})^{-1}\bm{C}
XA−kX=B\bm{XA} - k\bm{X} = \bm{B}Factor: X(A−kI)=B\bm{X}(\bm{A} - k\bm{I}) = \bm{B}, then post-multiplyX=B(A−kI)−1\bm{X} = \bm{B}(\bm{A} - k\bm{I})^{-1}

Common mistake

You can't factor out a number

AX+2X=(A+2)X\bm{AX} + 2\bm{X} = (\bm{A} + 2)\bm{X} is wrong, because A+2\bm{A} + 2 is not defined. Write 2X=2IX2\bm{X} = 2\bm{IX}, so AX+2X=(A+2I)X\bm{AX} + 2\bm{X} = (\bm{A} + 2\bm{I})\bm{X}.

And X\bm{X} must come out on the same side in both terms: AX+XB\bm{AX} + \bm{XB} cannot be factored at all.

Worked example

Solve AX+BX=C\bm{AX} + \bm{BX} = \bm{C}

  • Complex familiar
  • Technology active

Given

A=[1203−11012],B=[1−11−22010−2],C=[32223−1],\bm{A} = \begin{bmatrix} 1 & 2 & 0 \\ 3 & -1 & 1 \\ 0 & 1 & 2 \end{bmatrix}, \quad \bm{B} = \begin{bmatrix} 1 & -1 & 1 \\ -2 & 2 & 0 \\ 1 & 0 & -2 \end{bmatrix}, \quad \bm{C} = \begin{bmatrix} 3 & 2 \\ 2 & 2 \\ 3 & -1 \end{bmatrix},

find the matrix X\bm{X} such that AX+BX=C\bm{AX} + \bm{BX} = \bm{C}.

5 steps

Try it yourself first, then check the working one step at a time.

Method

  1. Factor X\bm{X} out on the side it appears.

  2. Check the bracketed matrix is non-singular.

  3. Multiply by its inverse on the same side.

Watch out for

  • Calculating C(A+B)−1\bm{C}(\bm{A} + \bm{B})^{-1}: the order is wrong, and here the product does not even exist.

  • Finding A−1C+B−1C\bm{A}^{-1}\bm{C} + \bm{B}^{-1}\bm{C}: inverses do not distribute over a sum.

A\bm{A}, B\bm{B} and X\bm{X} are 3×33 \times 3 matrices with XA+X=B\bm{XA} + \bm{X} = \bm{B}, and A+I\bm{A} + \bm{I} is non-singular. Which expression is X\bm{X}?

Choose one answer

Without technology

Rearrange the equation to X=…\bm{X} = \dots and say which side you multiply on. You can also be asked to find an inverse from an identity, such as A2−3A+2I=0\bm{A}^2 - 3\bm{A} + 2\bm{I} = \bm{0}, with no arithmetic beyond the identity.

With technology

Enter A\bm{A}, B\bm{B} and C\bm{C} in the matrix editor, then type the product in the right order, such as [A]−1[B][\bm{A}]^{-1}[\bm{B}] or [C]([A]+[B])−1[\bm{C}]([\bm{A}] + [\bm{B}])^{-1}. Write the matrix equation you solved before the answer.

Exam tip

Communicate the method

A Paper 2 answer of just a matrix risks losing the method marks. Show the rearrangement ("pre-multiplying by (A+B)−1(\bm{A} + \bm{B})^{-1}"), state that the inverted matrix is non-singular (give its determinant), then the calculator result. Check by substituting X\bm{X} back if time allows.

Summary

  • A square matrix of any order has an inverse exactly when its determinant is non-zero.

  • Find 3×33 \times 3 and larger inverses with technology, and check them by showing AA−1=I\bm{AA}^{-1} = \bm{I}.

  • If AB=kI\bm{AB} = k\bm{I} with k≠0k \neq 0, then A−1=1kB\bm{A}^{-1} = \frac{1}{k}\bm{B}.

  • AX=B⇒X=A−1B\bm{AX} = \bm{B} \Rightarrow \bm{X} = \bm{A}^{-1}\bm{B}; XA=B⇒X=BA−1\bm{XA} = \bm{B} \Rightarrow \bm{X} = \bm{BA}^{-1}.

  • Factor X\bm{X} out on the correct side, using I\bm{I} for numbers: AX+kX=(A+kI)X\bm{AX} + k\bm{X} = (\bm{A} + k\bm{I})\bm{X}.

Finished the lesson?

Try the 2 quick checks above and the lesson completes itself, or mark it complete now.

Lesson 2 of 80% read

Save your progress with a free account