Skip to content

Mathematics · Ch 3 — Matrices

Properties of Multiplication of Matrices

3.4.6

Properties of Multiplication of Matrices

Properties of Multiplication of Matrices

Matrix multiplication follows algebraic rules similar to — but not identical with — those for real numbers. The key difference is that it is not commutative: in general, AB≠BAAB \neq BA. However, three fundamental properties do hold whenever the products are defined.


1. The Associative Law

Associative Law: For any three matrices AA, BB and CC,

(AB)C=A(BC)(AB)C = A(BC)

whenever both sides are defined.

The grouping of the factors does not affect the result: you may multiply ABAB first, then by CC, or BCBC first, then pre-multiply by AA.

Note

The associative law is what lets us write ABCABC without parentheses — the product is unambiguous.


2. The Distributive Laws

Distributive Laws: For three matrices AA, BB and CC,

(i)A(B+C)=AB+AC(i)\quad A(B+C) = AB + AC

(ii)(A+B)C=AC+BC(ii)\quad (A+B)C = AC + BC

whenever both sides are defined.

Matrix multiplication distributes over addition, in both a left and a right form.

Watch out

Order matters. Since matrix multiplication is not commutative, A(B+C)=AB+ACA(B+C) = AB + AC is true, but (B+C)A=BA+CA(B+C)A = BA + CA is a different (equally true) statement. You cannot swap the order of factors when distributing.


3. The Existence of Multiplicative Identity

Multiplicative Identity: For every square matrix AA of order nn, there exists an identity matrix II of the same order such that

IA=AI=AIA = AI = A

The identity matrix II (1's on the main diagonal, 0's elsewhere) acts like the number 1: multiplying any matrix by the identity of compatible order leaves it unchanged.

Note

For non-square matrices the identity must match the dimensions. If AA is m×nm \times n, then ImA=AI_m A = A and AIn=AA I_n = A, with ImI_m and InI_n being different-sized identity matrices.


4. Application: Polynomials in Matrices

The properties above let us work with matrix polynomials. For a square matrix AA,

A2=A⋅A,A3=A2⋅A,and so onA^2 = A \cdot A, \quad A^3 = A^2 \cdot A, \quad \text{and so on}

with A0=IA^0 = I by convention. …