Properties of determinants
Determinants change in predictable ways under row operations and products. These properties make them fast to compute and powerful in theory.
Row operations
Adding a multiple of one row to another leaves the determinant unchanged. Swapping two rows multiplies it by -1. Multiplying one row by a scalar k multiplies the determinant by k. These three rules let us reduce a matrix to triangular form while tracking how the determinant changes, then multiply the diagonal.
Computing by row reduction
To find \(\det A\), row reduce to an echelon form U using row replacements and swaps only. Then \(\det A = (-1)^r \times\) (product of U's diagonal entries), where r is the number of swaps. This takes about \(\frac{2}{3}n^3\) operations instead of n!, making determinants of large matrices practical.
Invertibility
A square matrix is invertible if and only if \(\det A \ne 0\). Row reduction explains why: if A is singular, its echelon form has a zero on the diagonal, giving determinant zero; if invertible, all pivots are non-zero. This criterion joins the invertible matrix theorem.
Products and inverses
The determinant is multiplicative: \(\det(AB) = \det A \det B\). Consequently \(\det(A^{-1}) = \frac{1}{\det A}\), and \(\det(A^k) = (\det A)^k\). Note that determinants are not additive: in general \(\det(A + B) \ne \det A + \det B\).
Transposes and scalars
\(\det(A^T) = \det A\), so every row property also holds for columns: swapping columns flips the sign, and a matrix with two equal columns, or a zero column, has determinant zero. For an n x n matrix, \(\det(cA) = c^n\det A\), since each of the n rows is scaled by c.
Linearity in each row
The determinant is linear in each row separately when the other rows are held fixed: if one row is a sum u + v, the determinant splits into the sum of the determinants with u and with v in that row. Together with the swap rule and det I = 1, this property characterises the determinant uniquely.
Cramer's rule
For an invertible A, Gabriel Cramer's 1750 rule gives the solution of \(A\mathbf{x} = \mathbf{b}\) as \(x_i = \frac{\det A_i}{\det A}\), where \(A_i\) is A with column i replaced by \(\mathbf{b}\). It is elegant for 2 x 2 or 3 x 3 systems and useful in theory, but far too slow for large systems compared with elimination.
Worked example
Find \(\det\begin{bmatrix} 1 & 2 & 3 \\ 2 & 5 & 8 \\ 3 & 7 & 12 \end{bmatrix}\). Subtract 2 times row 1 from row 2 and 3 times row 1 from row 3: rows (1, 2, 3), (0, 1, 2), (0, 1, 3). Subtract row 2 from row 3: (0, 0, 1). No swaps were used, so the determinant is \(1 \times 1 \times 1 = 1\). Since it is non-zero, the matrix is invertible, and its inverse has integer entries.
Key takeaways
Row replacement leaves the determinant unchanged, a swap multiplies it by -1, and scaling a row by k multiplies it by k. Row reduce to triangular form and multiply the diagonal, adjusting for swaps: about two-thirds n^3 operations. A is invertible exactly when det A is not zero. det(AB) = det A det B, det(A^{-1}) = 1/det A, but determinants are not additive. det(A^T) = det A, so column rules mirror row rules; det(cA) = c^n det A. The determinant is linear in each row. Cramer's rule (1750) solves small systems with determinant ratios.
Flashcard
Determinant unchanged
Check your understanding
If A is a 3 x 3 matrix with det A = 5, what is det(2A)?
