PROJECTENGRAM

PROPERTIES OF MATRIX OPERATIONS

Properties of Matrix Operations

Tags: #◇CSMATH2 #◇Linear Algebra #◇Computational Mathematics #◇Matrices #◇Matrix Operations #◇Computer Science #◇Vector Spaces #◇Linear Transformations


◇Overview

Let A,B,CA,B,C be matrices of appropriate sizes, OO be the zero matrix, and r,sr,s be scalars.

◇Details / Theory

Let A,B,CA,B,C be matrices of appropriate sizes, OO be the zero matrix, and r,sr,s be scalars.

◇Matrix Addition Properties:

  1. Commutative: A+B=B+AA+B=B+A (Handout 1, Page 13)
  2. Associative: (A+B)+C=A+(B+C)(A+B)+C=A+(B+C) (Handout 1, Page 15)
  3. Additive Identity: A+O=AA+O=A (Handout 1, Page 16)
  4. Additive Inverse: For each AA, there exists a unique matrix −A-A such that A+(−A)=OA+(-A)=O. (Handout 1, Page 17)

◇Scalar Multiplication Properties:

  1. Associative: (rs)A=r(sA)(rs)A=r(sA) (Handout 1, Page 18)
  2. Distributive: r(A+B)=rA+rBr(A+B)=rA+rB and (r+s)A=rA+sA(r+s)A=rA+sA (Handout 1, Page 19)
  3. Scalar Identity/Zero: 1A=A1A=A, 0A=O0A=O, rO=OrO=O (Handout 1, Page 20)

◇Matrix Multiplication Properties:

  1. Associative: (AB)C=A(BC)(AB)C=A(BC) (Provided sizes are compatible) (Handout 1, Page 23)
    PN: Associativity means we can group multiplications, but we cannot change the order.

  2. Distributive: A(B+C)=AB+ACA(B+C)=AB+AC (Left) and (A+B)C=AC+BC(A+B)C=AC+BC (Right) (Provided sizes are compatible) (Handout 1, Page 24)

  3. Scalar Interaction: (rA)B=r(AB)=A(rB)(rA)B=r(AB)=A(rB) (Handout 1, Page 26)

◇Transpose of a Matrix Properties:

  1. Transpose of Sum: (A+B)T=AT+BT(A+B)^T=A^T+B^T (Handout 1, Page 29)

  2. Transpose of Product: (AB)T=BTAT(AB)^T=B^T A^T (Note the reversal of order!) (Handout 1, Page 29)
    PN: This reversal of order for the transpose of a product is extremely important and often tested.

  3. Transpose of Transpose: (AT)T=A(A^T)^T=A (Handout 1, Page 29)

  4. Transpose of Scalar Multiple: (rA)T=rAT(rA)^T=rA^T (Handout 1, Page 29)

◇Common Pitfalls

Matrix multiplication is NOT commutative: AB≠BAAB \neq BA in general.
(A+B)2≠A2+2AB+B2(A + B)^2 \neq A^2 + 2AB + B^2 unless AB=BAAB = BA.
(AB)T=BTAT(AB)^T = B^T A^T — note the reversed order.

◇Summary

TL;DR — Matrix operations are associative and distributive but multiplication is not commutative. Transpose reverses product order. These properties underpin all matrix algebra proofs.

Related: - ◇Matrix Addition - ◇Scalar Multiplication - ◇Matrix Multiplication - ◇Transpose of a Matrix
Source / Reference:

◇References & Navigation

CONFIRMATION // 理解度の確認

ACTIVE RECALL VERIFICATION

Select an answer to verify conceptual comprehension and reveal the technical rationale.

Q1

Which property asserts that for matrix addition, A+B=B+AA + B = B + A?

Q2

What is the additive identity matrix that satisfies A+O=AA + O = A?

Q3

What does the associative property of matrix multiplication mathematically state?

Q4

According to the distributive properties for matrices, what is the correct right-distributive expansion of (A+B)C(A+B)C?

Q5

How does a scalar rr interact within matrix multiplication (rA)B(rA)B?

Q6

What is the result of taking the transpose of a sum, (A+B)T(A + B)^T?

Q7

What is the crucial property to remember regarding the transpose of a matrix product, (AB)T(AB)^T?

Q8

What is the result of taking the transpose of a transpose, (AT)T(A^T)^T?

Q9

How is the transpose of a scalar multiple (rA)T(rA)^T correctly computed?

Q10

Which statement outlines a common pitfall regarding the binomial expansion (A+B)2(A+B)^2 for matrices?

Properties of Matrix Operations | PROJECT ENGRAM