DPST1013 Chap.5 Matrices, Inverses and Determinants
Matrices, Inverses and Determinants
Matrix algebra provides a compact language for linear systems and transformations. The dimensions of a matrix determine which operations are meaningful, and multiplication combines rows of one matrix with columns of another. Familiar scalar rules survive only in carefully stated forms: multiplication is associative and distributive, but the order of factors generally matters.
Inverses and determinants supply complementary information. An inverse, when it exists, undoes a square matrix's action and solves a system for any right-hand vector. A determinant is a single scalar that detects invertibility and records how row operations affect an oriented area or volume scale. Computing one does not mean computing the other.
This chapter compares direct matrix products, transpose identities, inverse methods and determinant calculations. The useful habits are to check dimensions before multiplying, preserve factor order, establish invertibility before cancellation, and keep a ledger of determinant changes during elimination. Each method has a natural verification that can expose a wrong sign or an unjustified algebraic step.
What this chapter covers
- 01
Matrix dimensions and row-column multiplication
- 02
Column combinations and the meaning of Ax
- 03
Transpose identities and reversed factor order
- 04
Two-by-two inverses and augmented reduction
- 05
Determinants, cofactors and row-operation effects
- 06
Equivalent tests for square-matrix invertibility
Use an inverse only after establishing it exists
- 1The determinant is 3·1−1·2=1, so A is invertible. The nonzero test must precede the inverse formula; a square shape alone would not establish that an inverse exists. Here the determinant is exactly one, so no fractional scale factor remains.
- 2Interchange the diagonal entries and negate the off-diagonal entries. The inverse therefore has first row (1,−1) and second row (−2,3). Multiplying A by this matrix gives rows (3−2,−3+3) and (2−2,−2+3), which are the identity rows (1,0) and (0,1).
- 2Apply the inverse to the right-hand column: the components are 8−5=3 and −16+15=−1. Substitute these values into the original equations, obtaining 3·3−1=8 and 2·3−1=5. Both equations are satisfied, verifying the solution independently of the inverse construction.
Key terms
- Conformable product
- A matrix product whose inner dimensions agree. Multiplying an m-by-n matrix by an n-by-p matrix produces an m-by-p matrix; changing the factor order may change the shape or make the product undefined.
- Transpose
- The matrix obtained by interchanging rows and columns. Transposition leaves a square determinant unchanged, but transposing a product requires reversing the factor order as well as transposing each individual factor.
- Identity matrix
- A square matrix with ones on the diagonal and zeros elsewhere. It leaves compatible vectors or matrices unchanged under multiplication and is the result required when a matrix is multiplied by its inverse.
- Inverse matrix
- A matrix that reverses the action of an invertible square matrix. Multiplying the original and inverse matrices gives the identity; an inverse cannot be assumed for a singular matrix or used as unspecified scalar division.
- Determinant
- A scalar defined for a square matrix. Its nonzero value characterises invertibility, while its magnitude and sign have geometric scaling and orientation interpretations in low-dimensional transformation examples.
- Cofactor
- A signed minor used in determinant expansion. The sign at row i and column j is determined by the alternating checkerboard pattern, so an omitted negative sign can spoil an otherwise correct minor calculation.
- Singular matrix
- A square matrix without an inverse, equivalently one with zero determinant. Its homogeneous system has a nonzero solution, and a particular nonhomogeneous right side may lead to either no solutions or infinitely many.
Matrices, Inverses and Determinants FAQ
Why can two valid products AB and BA have different answers?
Matrix multiplication represents ordered combinations of operations and uses different row-column pairings in each order. Even when both products have the same dimensions, their entries need not agree. Associativity permits regrouping factors; it does not permit exchanging their positions.
Does a zero determinant mean that every associated system has no solution?
No. It means the square coefficient matrix is singular. Some right-hand vectors may be attainable in infinitely many ways, while others are unattainable. Examine the augmented system for the particular right-hand vector to distinguish those two possibilities.
When should I use an inverse instead of Gaussian elimination?
An inverse can be efficient for a small invertible matrix or repeated right-hand vectors, especially when the inverse is already available. For a single larger system, direct elimination may require less work. In either case, establish the conditions and verify the final solution.
How does scaling a whole matrix affect its determinant?
If an n-by-n matrix is multiplied by c, each of its n rows is scaled by c, so the determinant is multiplied by c to the power n. Scaling only one row multiplies the determinant by c once, a different operation.
Why does the inverse of a product reverse the factor order?
To undo AB, the action of A must be undone first on the left, followed by undoing B. Algebraically, B inverse times A inverse makes the adjacent factors cancel correctly. The identity assumes both matrices are invertible and dimensionally compatible.
Exam move
Start each multiplication exercise by writing the input and expected output dimensions. Compute one entry slowly as a row-column dot product, then use that pattern for the remaining entries. Compare a small pair of products in both orders to build a concrete memory of noncommutativity. When practising determinants by row reduction, keep a separate line recording swaps and scaling factors.
Do not confuse preservation of a linear system's solutions with preservation of its determinant. Verify one reduced determinant through a cofactor expansion on the original matrix, preferably along a sparse row or column. For inverse problems, prove existence before using a formula, multiply the result back to the identity, and only then apply it to a right-hand vector.
Finally, connect singularity to the homogeneous system by finding a nonzero vector sent to zero in a simple example. This makes the abstract invertibility criteria easier to recognise during unfamiliar calculations.
Working through Matrices, Inverses and Determinants in DPST1013? Sia is AskSia’s AI Mathematics tutor — ask any DPST1013 Matrices, Inverses and Determinants question and get a clear, step-by-step explanation grounded in how DPST1013 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.
