A Matrix Is a Linear Transformation

I. The Core Idea: Recording Where the Basis Vectors Go

The previous section said that once you know where the basis vectors ı^\hat{\imath} and ȷ^\hat{\jmath} go, the whole linear transformation is determined. A matrix is the table that records this “destination”.

Suppose the linear transformation TT satisfies:

T(ı^)=(ac),T(ȷ^)=(bd)T(\hat{\imath}) = \begin{pmatrix} a \\ c \end{pmatrix}, \quad T(\hat{\jmath}) = \begin{pmatrix} b \\ d \end{pmatrix}

then record it with a 2×2 matrix:

M=(abcd)M = \begin{pmatrix} a & b \\ c & d \end{pmatrix}
  • Column 1 (a,c)T(a, c)^T = the new position of ı^\hat{\imath}
  • Column 2 (b,d)T(b, d)^T = the new position of ȷ^\hat{\jmath}

Each column of a matrix is the image of the corresponding basis vector.

II. Matrix Times Vector: How to Get the Image of Any Vector

For an arbitrary vector v=(x,y)T=xı^+yȷ^\mathbf{v} = (x, y)^T = x\hat{\imath} + y\hat{\jmath}, linearity gives:

T(v)=x T(ı^)+y T(ȷ^)=x(ac)+y(bd)=(ax+bycx+dy)T(\mathbf{v}) = x\,T(\hat{\imath}) + y\,T(\hat{\jmath}) = x\begin{pmatrix} a \\ c \end{pmatrix} + y\begin{pmatrix} b \\ d \end{pmatrix} = \begin{pmatrix} ax + by \\ cx + dy \end{pmatrix}

This is the definition of matrix multiplication:

 Mv=(abcd)(xy)=(ax+bycx+dy) \boxed{\,M\mathbf{v} = \begin{pmatrix} a & b \\ c & d \end{pmatrix}\begin{pmatrix} x \\ y \end{pmatrix} = \begin{pmatrix} ax + by \\ cx + dy \end{pmatrix}\,}

The rule of computation

“Row times column”: the ii-th component of the result = the ii-th row of the matrix multiplied component by component with the vector and then summed.

(abcd)(xy)=(ax+bycx+dy)\begin{pmatrix} a & b \\ c & d \end{pmatrix}\begin{pmatrix} x \\ y \end{pmatrix} = \begin{pmatrix} ax + by \\ cx + dy \end{pmatrix}
Example: M = [[2, 0], [0, 0.5]] (x stretched by 2, y compressed by half)
M·(1, 1) = (2×1 + 0×1, 0×1 + 0.5×1) = (2, 0.5)

III. Watching a Matrix “Move” Space

Key observations:

  • column 1 of the matrix (a,c)(a, c) = the new position of the blue basis vector ı^\hat{\imath}
  • column 2 of the matrix (b,d)(b, d) = the new position of the red basis vector ȷ^\hat{\jmath}
  • change the matrix entries → the grid changes with it → a matrix and a transformation are two ways of writing the same thing

IV. Some Common Matrices and Their Geometric Meanings

MatrixTransformationGeometric effect
(1001)\begin{pmatrix} 1 & 0 \\ 0 & 1 \end{pmatrix}the identity transformationnothing changes
(0−110)\begin{pmatrix} 0 & -1 \\ 1 & 0 \end{pmatrix}a 90° rotationa quarter turn counterclockwise
(2000.5)\begin{pmatrix} 2 & 0 \\ 0 & 0.5 \end{pmatrix}a scalingx stretched by 2, y compressed by half
(1101)\begin{pmatrix} 1 & 1 \\ 0 & 1 \end{pmatrix}a horizontal shearthe vertical direction is unchanged, the horizontal is offset by y
(−1001)\begin{pmatrix} -1 & 0 \\ 0 & 1 \end{pmatrix}a reflectiona mirror image about the y-axis

Verifying the rotation matrix: ı^=(1,0)→(0,1)\hat{\imath} = (1,0) \to (0,1), ȷ^=(0,1)→(−1,0)\hat{\jmath} = (0,1) \to (-1,0) — after a counterclockwise rotation by 90°, the positive x-axis turns into the positive y-axis, and the positive y-axis turns into the negative x-axis ✓

V. Why Column 1 Is the Image of ı^\hat{\imath}

Substitute ı^=(1,0)T\hat{\imath} = (1, 0)^T into the matrix multiplication:

Mı^=(abcd)(10)=(a⋅1+b⋅0c⋅1+d⋅0)=(ac)M\hat{\imath} = \begin{pmatrix} a & b \\ c & d \end{pmatrix}\begin{pmatrix} 1 \\ 0 \end{pmatrix} = \begin{pmatrix} a\cdot 1 + b\cdot 0 \\ c\cdot 1 + d\cdot 0 \end{pmatrix} = \begin{pmatrix} a \\ c \end{pmatrix}

It picks out exactly column 1. Similarly Mȷ^M\hat{\jmath} picks out column 2. This is the algebraic proof of “column = the image of a basis vector”.

Summary

Symbols Used in This Article

SymbolTypePronunciation/ExplanationMeaning in This Article
ı^,ȷ^\hat{\imath}, \hat{\jmath}Mathematical symboli hat, j hatThe unit basis vectors of the two-dimensional plane
T(ı^),T(ȷ^)T(\hat{\imath}), T(\hat{\jmath})Mathematical symbolT of i-hat, T of j-hatThe images of the basis vectors, that is, the columns of the matrix
M=(abcd)M = \begin{pmatrix} a & b \\ c & d \end{pmatrix}Mathematical symbol2 by 2 matrixThe 2×2 matrix recording the linear transformation
MvM\mathbf{v}Mathematical symbolM times vThe matrix times a vector, giving (ax+bycx+dy)\begin{pmatrix} ax+by \\ cx+dy \end{pmatrix}
(a,c)T(a, c)^TMathematical symbola c transposeColumn 1 of the matrix, the image of ı^\hat{\imath}
v=(x,y)T\mathbf{v} = (x, y)^TMathematical symbolvector vAn arbitrary input vector

Found an issue with this content?

Tell us what could be improved.