Matrix Calculator

A matrix calculator performs linear algebra operations on rectangular arrays of numbers.

From simple addition to complex operations like finding inverses and row-reducing, our calculator handles all standard matrix operations with full step-by-step explanations.

• Addition/subtraction requires equal dimensions and works element-wise.
• Multiplication requires A's columns = B's rows.
• The determinant is a scalar that indicates invertibility.
• The inverse undoes multiplication (A × A⁻¹ = I).
• RREF uses Gauss-Jordan elimination to solve linear systems.

A matrix is a rectangular array of numbers arranged in rows and columns, written inside brackets and sized by its dimensions as rows-by-columns (for example, a 2×3 matrix has 2 rows and 3 columns). Each entry is referenced by its position, A[i][j], meaning the number in row i and column j.

According to Encyclopaedia Britannica, matrices formalize systems of linear equations and linear transformations into a single compact object.

A square matrix has equal rows and columns, and its main diagonal runs from top-left to bottom-right. The identity matrix I has 1s on that diagonal and 0s elsewhere, playing the role that the number 1 plays in ordinary multiplication.

To multiply matrices A and B, each entry C[i][j] is the dot product of row i of A with column j of B, summing A[i][k]×B[k][j] across k. This is only defined when the number of columns in A equals the number of rows in B, so an m×n matrix times an n×p matrix yields an m×p result.

For example, [[1,2],[3,4]] × [[5,6],[7,8]] gives C[1][1] = 1×5 + 2×7 = 19, and completing the process produces [[19,22],[43,50]].

Khan Academy stresses that matrix multiplication is not commutative: A×B rarely equals B×A. This row-by-column rule underlies how linear transformations compose.

The determinant of a 2×2 matrix [[a,b],[c,d]] equals ad − bc; for [[1,2],[3,4]] that is (1×4) − (2×3) = 4 − 6 = −2.

For a 3×3 matrix you expand along a row or column using cofactors, multiplying each entry by the determinant of the 2×2 minor left after deleting its row and column, alternating signs.

Wolfram MathWorld defines the determinant as a scalar encoding how a linear map scales signed area or volume. A determinant of zero means the matrix is singular (non-invertible) because it collapses space onto a lower dimension. Larger determinants are found efficiently through row reduction.

The inverse A⁻¹ is the matrix satisfying A × A⁻¹ = I, and it exists only when the determinant is nonzero.

For a 2×2 matrix [[a,b],[c,d]], the inverse is (1/det) × [[d,−b],[−c,a]]. Taking [[1,2],[3,4]] with determinant −2, the inverse is (1/−2) × [[4,−2],[−3,1]] = [[−2,1],[1.5,−0.5]].

For larger matrices, augment A with the identity to form [A | I] and apply Gauss-Jordan elimination until the left block becomes I, leaving A⁻¹ on the right.

Wolfram MathWorld notes the inverse directly solves Ax = b as x = A⁻¹b, which is why invertibility is central to linear systems.

Gaussian elimination solves linear systems by applying three elementary row operations until the matrix reaches row echelon form:

• swapping rows
• scaling a row by a nonzero constant
• adding a multiple of one row to another

Continuing until every pivot is 1 and is the only nonzero entry in its column produces the Reduced Row Echelon Form (RREF), the goal of Gauss-Jordan elimination.

Khan Academy explains that RREF makes the solution set of a system immediately readable. Because these operations never change that solution set, RREF is the most reliable general method for solving equations, computing rank, and inverting matrices, and it is numerically stable when combined with partial pivoting.

The rank of a matrix is the number of linearly independent rows, which equals the number of linearly independent columns and the count of pivot positions in its RREF.

According to Wolfram MathWorld, rank measures the dimension of the space spanned by the matrix's rows or columns.

A square n×n matrix is invertible exactly when it has full rank n; any lower rank signals a singular matrix and a determinant of zero. Rank also reveals how many equations in a system carry genuinely new information versus being redundant combinations of others.

In data science, low rank indicates that columns are highly correlated, which underpins techniques like principal component analysis and matrix compression.

Matrices power a remarkable range of technologies.

In computer graphics and games, transformation matrices rotate, scale, and project 3D scenes onto a 2D screen; multiplying these matrices composes several transformations into one.

In machine learning, the weights of neural networks are matrices, and training relies on massive matrix multiplications accelerated by GPUs.

A single motion relationship like distance = speed × time is simple enough for a speed distance time calculator, but Encyclopaedia Britannica notes that matrices take over once many quantities couple together in physics and economics, as with electrical circuits and input-output economic models.

Engineers use them for structural analysis and finite-element methods, while statisticians use covariance matrices to describe relationships between variables.

Google's original PageRank algorithm even framed the entire web as one enormous matrix eigenvalue problem.

An eigenvector of a square matrix A is a nonzero vector v whose direction is unchanged when multiplied by A, so that A·v = λ·v, where the scalar λ is the corresponding eigenvalue.

Wolfram MathWorld describes eigenvalues as the roots of the characteristic polynomial det(A − λI) = 0.

Geometrically, eigenvectors mark the axes along which a linear transformation simply stretches or compresses space by the factor λ, without rotating that direction.

Eigenvalues underpin stability analysis, vibration modes in engineering, quantum energy levels, and dimensionality reduction such as principal component analysis. The product of all eigenvalues equals the determinant, and their sum equals the trace (the sum of the diagonal entries).

• The most frequent error is assuming matrix multiplication is commutative — A×B usually does not equal B×A, and reversing the order gives a different or undefined result.
• Another is attempting to multiply matrices whose inner dimensions do not match; A's column count must equal B's row count.
• Students often try to invert a matrix with a zero determinant, which is impossible because such a matrix is singular.
• When transposing a product, remember the order reverses: (AB)ᵀ = BᵀAᵀ.
• During row reduction, forgetting to apply an operation to every entry in the row, or dividing by a zero pivot, corrupts the result.
• Finally, mixing up row and column indices when reading A[i][j] silently produces wrong answers.