Identity, Inverse, and Determinant Methods

The Central Idea

For an ordinary nonzero number \(a\), the reciprocal

satisfies

A matrix inverse works similarly.

If a square matrix \(A\) has an inverse \(A^{-1}\), then

The identity matrix \(I\) plays the role of \(1\).

But not every square matrix has an inverse.

The determinant tells us when an inverse exists:

The Identity Matrix

An identity matrix is a square matrix with \(1\)s on the main diagonal and \(0\)s elsewhere.

For example,

and

Why the Identity Matrix Matters

For a compatible matrix \(A\),

and

The identity matrix leaves a matrix or vector unchanged under multiplication.

Identity as an Unchanged Transformation

Applied to a vector,

Geometrically, the identity transformation leaves every point where it is.

What Is an Inverse Matrix?

A square matrix \(A\) is invertible if there exists a matrix \(A^{-1}\) such that

The matrix \(A^{-1}\) is called the inverse of \(A\).

Inverse Means Undoing

If \(A\) transforms a vector

into

then

The inverse reverses the action of \(A\).

Not Every Matrix Is Invertible

Consider

The second row is twice the first.

The matrix contains dependent information.

It cannot be inverted.

Singular and Nonsingular

A square matrix with no inverse is called singular.

A square matrix with an inverse is called nonsingular or invertible.

The Determinant

The determinant is a scalar associated with a square matrix.

It is written

or sometimes

The determinant contains information about invertibility, scaling, orientation, and dependence.

Determinant of a \(2\times2\) Matrix

For

the determinant is

Example: \(2\times2\) Determinant

Let

Then

Determinant Zero Means No Inverse

For

we have

Therefore,

Determinant Nonzero Means Invertible

For a square matrix \(A\),

This is one of the central tests for invertibility.

The \(2\times2\) Inverse Formula

If

and

then

How to Remember the \(2\times2\) Formula

Starting with

• swap the main-diagonal entries \(a\) and \(d\);
• change the signs of \(b\) and \(c\);
• divide by the determinant.

Example: Find an Inverse

Let

We already found

Therefore,

Simplify the Inverse

Verify an Inverse

To verify \(A^{-1}\), compute

For

and

the product is

Why Inverse Order Matters

Matrix multiplication is generally not commutative.

For an inverse, both conditions hold:

and

This two-sided property is part of what makes \(A^{-1}\) the inverse.

Inverse of the Identity Matrix

The identity matrix is its own inverse:

Inverse of an Inverse

If \(A\) is invertible, then

Inverse of a Product

If \(A\) and \(B\) are invertible, then

The order reverses.

Why the Order Reverses

Check:

The last transformation applied must be the first one undone.

Solving \(A\mathbf{x}=\mathbf{b}\) with an Inverse

Suppose

and \(A\) is invertible.

Multiply both sides on the left by \(A^{-1}\):

Then

so

Example: Solve a System with an Inverse

Consider

Write

Then

Compute the Inverse Solution

Using

we obtain

So

Therefore,

Interpret the Vector Solution

The vector equation gives

Check:

Why the Inverse Method Requires a Square Matrix

Ordinary matrix inverses are defined for square matrices.

A rectangular coefficient matrix may still describe a valid system, but the simple formula

does not apply.

Why the Inverse Method Requires Invertibility

If

then \(A^{-1}\) does not exist.

The system may then have:

• no solution;
• infinitely many solutions.

Row reduction is needed to distinguish the cases.

Gauss–Jordan Method for Finding an Inverse

For an invertible \(n\times n\) matrix \(A\), augment it with the identity matrix:

Then use Gauss–Jordan elimination to transform the left side into \(I\).

If successful, the right side becomes \(A^{-1}\):

Example: Start Gauss–Jordan Inversion

Let

Begin with

Swap for a Convenient Pivot

Use

Then

Eliminate Below the Pivot

Use

Then

Scale the Second Pivot

Use

Then

Clear Above the Second Pivot

Use

Then

Therefore,

How Gauss–Jordan Detects Noninvertibility

If the left side of

cannot be reduced to the identity because a pivot is missing, then \(A\) is singular.

No inverse exists.

Determinant of a \(3\times3\) Matrix

For

one cofactor expansion along the first row gives

Example: \(3\times3\) Determinant

Let

Expand along the first row:

Finish the \(3\times3\) Determinant

So

Because the determinant is nonzero, \(A\) is invertible.

Minors and Cofactors Preview

The determinant formulas for larger matrices use minors and cofactors.

For an entry \(a_{ij}\):

• delete row \(i\) and column \(j\);
• take the determinant of the remaining matrix to obtain the minor \(M_{ij}\);
• apply the sign factor \((-1)^{i+j}\) to obtain the cofactor.

Cofactor Sign Pattern

For a \(3\times3\) matrix, the signs are

Cofactor Expansion

A determinant can be expanded along any row or column.

Choosing a row or column with many zeros usually reduces the amount of arithmetic.

Triangular Matrices Have Easy Determinants

If \(A\) is triangular, then

For example,

Determinant of the Identity Matrix

Determinant of a Product

For square matrices of the same size,

Determinant of an Inverse

If \(A\) is invertible, then

Therefore,

Determinant and Transpose

Transposing a square matrix does not change its determinant.

Effect of Row Swapping

Swapping two rows changes the sign of the determinant:

if \(B\) is obtained by one row swap.

Effect of Row Scaling

If one row is multiplied by \(k\), the determinant is multiplied by \(k\).

Thus,

Effect of Row Replacement

Replacing one row by itself plus a multiple of another row does not change the determinant.

This makes elimination useful for determinant computation.

Determinant by Elimination

For a large matrix, it is often efficient to row reduce toward triangular form while tracking determinant changes.

Once triangular, multiply the diagonal entries and adjust for row swaps and row scalings.

Determinant and Row Dependence

If one row is a scalar multiple of another, then

More generally, linearly dependent rows or columns force determinant zero.

Repeated Rows Force Determinant Zero

If two rows are identical, swapping them leaves the matrix unchanged.

But a row swap should negate the determinant.

The only scalar equal to its own negative is zero.

So the determinant must be zero.

Determinant as Area Scaling

For a \(2\times2\) transformation matrix \(A\),

is the factor by which areas are scaled.

Example: Area Scaling

Let

Then

Any planar region has its area multiplied by

Determinant as Volume Scaling

For a \(3\times3\) transformation matrix,

gives the volume-scaling factor.

Negative Determinants and Orientation

The absolute value of the determinant measures scale.

The sign also records orientation.

A negative determinant indicates an orientation reversal, such as a reflection combined with scaling.

Why Determinant Zero Means Collapse

If

the transformation collapses dimension.

For example, a two-dimensional region may be flattened into a line.

Once information is collapsed in this way, no inverse transformation can recover the original input uniquely.

Invertibility and Unique Solutions

For a square coefficient matrix \(A\), the following ideas are connected:

• \(A\) is invertible;
• \(\det(A)\ne0\);
• every variable column has a pivot;
• the RREF of \(A\) is \(I\);
• \(A\mathbf{x}=\mathbf{b}\) has a unique solution for every \(\mathbf{b}\).

The Invertible Matrix Theorem Preview

Later linear algebra collects many equivalent conditions for invertibility into the Invertible Matrix Theorem.

The equivalences above are among its most important components.

Cramer’s Rule

For a square system

with

Cramer’s Rule expresses each variable as a ratio of determinants.

Cramer’s Rule for Two Variables

For

let

Then

and

Example: Cramer’s Rule

Solve

The coefficient determinant is

Compute \(D_x\)

Replace the \(x\)-column with constants:

Therefore,

Compute \(D_y\)

Replace the \(y\)-column with constants:

Thus,

The solution is

When Cramer’s Rule Applies

Cramer’s Rule requires:

• a square coefficient matrix;
• a nonzero determinant.

If

the formulas divide by zero and cannot be used.

Cramer’s Rule Is Elegant but Not Usually Efficient for Large Systems

For small systems, Cramer’s Rule is conceptually useful.

For large systems, repeated determinant computation becomes expensive.

Gaussian elimination is generally more efficient.

Inverse Method Versus Cramer’s Rule

Both require an invertible square coefficient matrix.

The inverse method uses

Cramer’s Rule uses determinant ratios.

They are different expressions of the same unique-solvability structure.

Inverse Method Versus Gaussian Elimination

If solving only one system, explicitly computing \(A^{-1}\) may require more work than Gaussian elimination.

If the same matrix \(A\) is used with many different right-hand sides, an already-computed inverse can be reused:

But Numerical Computation Usually Avoids Explicit Inverses

In practical numerical computing, solving

through factorization or elimination is often more stable and efficient than explicitly forming \(A^{-1}\).

The inverse remains conceptually important even when it is not computed directly.

Determinants and Numerical Scale

A determinant close to zero may indicate that a matrix is nearly singular.

Such systems can be sensitive to small measurement or rounding errors.

Later numerical methods study this sensitivity more carefully through conditioning.

Small Determinant Does Not Alone Measure Conditioning

Although a very small determinant can be a warning sign in some contexts, determinant magnitude by itself is not a complete measure of numerical stability.

Matrix scaling and dimension also matter.

Condition numbers provide a better later tool.

Determinants Apply Only to Square Matrices

Ordinary determinants are defined for square matrices.

A rectangular matrix may still have rank, pivots, and important linear structure, but it does not have a determinant in the ordinary sense.

Inverse Matrices Apply Only to Square Matrices in This Sense

A rectangular matrix does not have a two-sided inverse satisfying

and

with the same ordinary inverse concept.

Later mathematics develops generalized inverses for rectangular matrices.

Applied Transformation Example

Suppose

scales horizontal coordinates by \(2\) and vertical coordinates by \(3\).

The inverse is

The inverse undoes the scaling.

Check the Transformation Inverse

Applied Calibration Example

Suppose two measured outputs depend linearly on two hidden quantities:

If \(A\) is invertible, then

The inverse converts measurements back into the underlying quantities.

Invertibility Means Information Is Recoverable

An invertible linear transformation keeps enough information to reconstruct the input uniquely.

A singular transformation loses at least one independent direction of information.

Determinant and Parallel Columns

For a \(2\times2\) matrix, if the columns point in the same or opposite direction, they span no area.

The determinant is zero.

This geometric dependence matches algebraic noninvertibility.

Determinant and Basis Preview

A collection of \(n\) column vectors in \(\mathbb{R}^n\) forms a basis exactly when the square matrix made from those columns has nonzero determinant.

This connection will later link determinants with linear independence and coordinate systems.

A Reliable \(2\times2\) Determinant Workflow

1. Identify \(a,b,c,d\).
2. Compute \(ad\).
3. Compute \(bc\).
4. Subtract: \(ad-bc\).
5. Use the sign and zero/nonzero status appropriately.

A Reliable \(2\times2\) Inverse Workflow

1. Compute \(\det(A)=ad-bc\).
2. If the determinant is zero, stop: no inverse exists.
3. Swap \(a\) and \(d\).
4. Negate \(b\) and \(c\).
5. Multiply by \(1/\det(A)\).
6. Verify by multiplication if needed.

A Reliable Gauss–Jordan Inverse Workflow

1. Form \([A\mid I]\).
2. Use Gauss–Jordan elimination on the entire augmented matrix.
3. Attempt to turn the left side into \(I\).
4. If successful, read \(A^{-1}\) on the right.
5. If the left side cannot become \(I\), \(A\) is singular.

A Reliable Inverse-System Workflow

1. Write the system as \(A\mathbf{x}=\mathbf{b}\).
2. Check that \(A\) is square.
3. Check that \(A\) is invertible.
4. Compute \(A^{-1}\) if appropriate.
5. Evaluate \(\mathbf{x}=A^{-1}\mathbf{b}\).
6. Verify the solution in the original system.

A Reliable Cramer’s Rule Workflow

1. Form the coefficient matrix \(A\).
2. Compute \(D=\det(A)\).
3. If \(D=0\), Cramer’s Rule does not apply.
4. Replace one variable column at a time with the constant vector.
5. Compute the replacement determinants.
6. Divide each replacement determinant by \(D\).
7. Verify the resulting solution.

Common Errors

Using an Inverse Formula When the Determinant Is Zero

Division by \(\det(A)\) is impossible when the determinant is zero.

Confusing Entrywise Reciprocals with a Matrix Inverse

\(A^{-1}\) is not obtained by taking \(1/a_{ij}\) entry by entry.

Forgetting the Sign Changes in the \(2\times2\) Formula

The off-diagonal entries are negated.

Multiplying on the Wrong Side

From \(A\mathbf{x}=\mathbf{b}\), multiply by \(A^{-1}\) on the left.

Assuming Every Square Matrix Is Invertible

Square is necessary, but nonzero determinant is also required.

Using Determinants on Rectangular Matrices

Ordinary determinants require square matrices.

Forgetting Row-Operation Effects on Determinants

Row swaps negate, row scaling multiplies, and row replacement preserves the determinant.

Using Cramer’s Rule When \(D=0\)

The determinant ratios are undefined in that case.

Assuming a Small Determinant Alone Measures Numerical Instability

Conditioning requires a broader analysis.

Computing an Explicit Inverse When Elimination Would Be Simpler

Choose a method that matches the problem rather than forcing one technique.

Practice

1. Identity Matrix

Write the \(3\times3\) identity matrix.

2. Compute a Determinant

Find

3. Test Invertibility

Is

invertible?

4. Find a \(2\times2\) Inverse

Find the inverse of

5. Verify an Inverse

Verify that

is the inverse of

6. Solve by Inverse

Solve

7. Triangular Determinant

Find

8. Row Swap Effect

If \(\det(A)=7\) and \(B\) is obtained from \(A\) by swapping two rows once, what is \(\det(B)\)?

9. Row Scaling Effect

If \(\det(A)=5\) and one row is multiplied by \(3\), what happens to the determinant?

10. Cramer’s Rule

Solve

using Cramer’s Rule.

11. Area Scaling

A transformation matrix has determinant

By what factor are areas scaled?

12. Invertibility from RREF

A square matrix reduces to the identity matrix. What can you conclude?

Key Ideas

• The identity matrix acts as the multiplicative identity for matrices.
• An inverse matrix satisfies \(AA^{-1}=A^{-1}A=I\).
• An inverse reverses the action of an invertible linear transformation.
• Not every square matrix is invertible.
• A square matrix is invertible exactly when its determinant is nonzero.
• For a \(2\times2\) matrix, \(\det(A)=ad-bc\).
• The \(2\times2\) inverse formula divides an adjusted matrix by the determinant.
• Gauss–Jordan elimination can compute an inverse through \([A\mid I]\to[I\mid A^{-1}]\).
• If \(A\) is invertible, \(A\mathbf{x}=\mathbf{b}\) has the unique solution \(\mathbf{x}=A^{-1}\mathbf{b}\).
• The determinant of a triangular matrix is the product of its diagonal entries.
• Row swaps negate determinants, row scaling scales them, and row replacement leaves them unchanged.
• \(|\det(A)|\) measures area or volume scaling for linear transformations.
• A zero determinant corresponds to dimensional collapse and loss of invertibility.
• Cramer’s Rule solves square invertible systems through ratios of determinants.
• Inverse, determinant, pivot, and unique-solution conditions are deeply connected.

Where This Leads

Identity, inverse, and determinant methods reveal a deeper structure behind systems of equations. An invertible matrix preserves enough information to recover inputs uniquely, while a singular matrix collapses at least one independent direction. The determinant detects that difference with a single scalar.

From here, the section can develop rank, homogeneous systems, linear independence, matrix transformations, determinant applications, and broader invertibility criteria. These ideas will later support eigenvalues, change of basis, geometry, numerical methods, optimization, differential systems, and many other areas of mathematics.