Skip to content

Determinant

Simple Idea

  • Think of the determinant of a square matrix as a scalar value that tells you about the scaling factor of the linear transformation represented by that matrix.
  • Specifically, it tells you how much the area (in 2D) or volume (in 3D) changes when you apply the matrix transformation. A determinant of 0 means the transformation squashes space into a lower dimension (like flattening a 3D object onto a plane or a 2D shape onto a line).

Formal Definition

  • The determinant is a scalar value that can only be computed for a square matrix \(\mathbf{A}\) (\(n \times n\)). It is denoted as \(\det(\mathbf{A})\) or \(|\mathbf{A}|\).
  • Its calculation encodes properties related to the matrix's invertibility and the geometric scaling effect of the corresponding linear transformation.

Key Concepts

1. Calculation

  • For a 2x2 Matrix: If \(\mathbf{A} = \begin{bmatrix} a & b \\ c & d \end{bmatrix}\), then: $$ \det(\mathbf{A}) = |\mathbf{A}| = ad - bc $$
  • For a 3x3 Matrix: Using cofactor expansion (e.g., along the first row): $$ |\mathbf{A}| = a \begin{vmatrix} e & f \ h & i \end{vmatrix} - b \begin{vmatrix} d & f \ g & i \end{vmatrix} + c \begin{vmatrix} d & e \ g & h \end{vmatrix} $$ $$ |\mathbf{A}| = a(ei - fh) - b(di - fg) + c(dh - eg) $$ (Where \(\begin{vmatrix} \dots \end{vmatrix}\) indicates the determinant of the 2x2 submatrix).
  • For Larger Matrices: Cofactor expansion can be used recursively, but it becomes computationally very expensive (\(\mathcal{O}(n!)\)). Methods based on row reduction (Gaussian elimination) to reach triangular form are more practical (\(\mathcal{O}(n^3)\)):
    • Swapping two rows multiplies the determinant by -1.
    • Multiplying a row by a scalar \(k\) multiplies the determinant by \(k\).
    • Adding a multiple of one row to another does not change the determinant.
    • The determinant of a triangular matrix (upper or lower) is the product of its diagonal entries.

2. Geometric Interpretation

  • \(|\det(\mathbf{A})|\) represents the factor by which area (in 2D) or volume (in 3D) or hypervolume (in higher dimensions) is scaled when applying the linear transformation defined by \(\mathbf{A}\).
  • Sign of the Determinant:
    • \(\det(\mathbf{A}) > 0\): Transformation preserves orientation (e.g., doesn't flip shapes inside-out).
    • \(\det(\mathbf{A}) < 0\): Transformation reverses orientation (e.g., like looking in a mirror).
    • \(\det(\mathbf{A}) = 0\): Transformation collapses space into a lower dimension (area/volume becomes zero). The matrix is singular (non-invertible).

3. Properties of Determinants

  • \(\det(\mathbf{A}) \neq 0\) if and only if \(\mathbf{A}\) is invertible.
  • \(\det(\mathbf{I}) = 1\) (The Identity matrix doesn't scale volume or change orientation).
  • \(\det(\mathbf{A}^T) = \det(\mathbf{A})\) (Transpose has the same determinant).
  • \(\det(\mathbf{AB}) = \det(\mathbf{A}) \det(\mathbf{B})\) (Determinant of a product is the product of determinants). This reflects composing scaling factors of transformations. Requires \(\mathbf{A}, \mathbf{B}\) to be square matrices of the same size.
  • \(\det(\mathbf{A}^{-1}) = 1 / \det(\mathbf{A}) = (\det(\mathbf{A}))^{-1}\) (if \(\mathbf{A}\) is invertible).
  • \(\det(k\mathbf{A}) = k^n \det(\mathbf{A})\) for an \(n \times n\) matrix \(\mathbf{A}\) and scalar \(k\).

Connections to Other Topics

  • Crucial for determining if a Matrix Inverse exists.
  • Used in Cramer's rule for solving systems of linear equations \(\mathbf{Ax=b}\) (though often not computationally preferred).
  • Appears in the change of variables formula for multivariable calculus integration (Jacobian determinant).
  • Related to Eigenvalues (the determinant is the product of the eigenvalues, counting algebraic multiplicities).

Summary

  • The Determinant (\(\det(\mathbf{A})\) or \(|\mathbf{A}|\)) is a scalar value associated with a square matrix.
  • Geometrically represents the volume scaling factor (and orientation change) of the associated linear transformation.
  • \(\det(\mathbf{A}) = 0 \iff\) the matrix collapses space \(\iff\) matrix is singular (non-invertible).
  • \(\det(\mathbf{A}) \neq 0 \iff\) matrix is invertible.
  • Key Property: \(\det(\mathbf{AB}) = \det(\mathbf{A})\det(\mathbf{B})\).

Sources