Skip to main content

Section 1.3 Vector Arithmetic

Subsection 1.3.1 Linear Operations

In this section, I go through the basic arithmetic of vectors. In some ways, it mimics the arithmetic of scalars; in other ways, scalar operations no longer work and are replaced with other operations.

Definition 1.3.1.

The sum of two vectors \(u\) and \(v\) in \(\RR^n\) is the sum taken component wise.
\begin{equation*} u + v = \begin{pmatrix} u_1 \\ u_2 \\ \vdots \\ u_n \end{pmatrix} + \begin{pmatrix} v_1 \\ v_2 \\ \vdots \\ v_n \end{pmatrix} = \begin{pmatrix} u_1 + v_1 \\ u_2 + v_2 \\ \vdots \\ u_n + v_n \end{pmatrix} \end{equation*}
The sum is visualized by placing the start of the second vector at the end of the first, as in Figure 1.3.2. Note that I can only add two vectors in the same dimension. I can’t add a vector in \(\RR^2\) to a vector in \(\RR^3\text{.}\)
Figure 1.3.2. Visualizing Vector Addition

Definition 1.3.3.

If \(u\) is a vector in \(\RR^n\) and \(a \in \RR\) is a real number, then the scalar multiplication of \(u\) and \(a\) is multiplication by \(a\) in each component of \(u\text{.}\) By convention, scalar multiplication is written with the scalar on the left of the vector.
\begin{equation*} au = a \begin{pmatrix} u_1 \\ u_2 \\ \vdots \\ u_n \end{pmatrix} = \begin{pmatrix} au_1 \\ au_2 \\ \vdots \\ au_n \end{pmatrix} \end{equation*}
Though there will be other ‘multiplications’ to come, it is mostly true that I can’t multiply vectors together in any way reminiscent of numbers. Instead, I can only multiply by scalars. Scalar multiplication is visualized by scaling the vector by the value of the scalar. (Hence the term ‘scalar’!) If the scalar is negative, the direction is also reversed, as in Figure 1.3.4.
Figure 1.3.4. Visualizing Scalar Multiplication
Scalar multiplication also lets me define the difference between vectors.

Definition 1.3.5.

The difference between two vectors \(u\) and \(v\) is the vector \(u + (-1)v\text{,}\) defined using addition and scalar multiplication. This works out to be componentwise subtraction.
\begin{equation*} u - v = u + (-1) v= \begin{pmatrix} u_1 \\ u_2 \\ \vdots \\ u_n \end{pmatrix} + (-1) \begin{pmatrix} v_1 \\ v_2 \\ \vdots \\ v_n \end{pmatrix} = \begin{pmatrix} u_1 - v_ 1\\ u_2 - v_2 \\ \vdots \\ u_n - v_n \end{pmatrix} \end{equation*}

Definition 1.3.6.

With respect to some set of scalars (such as \(\RR\)), whenever I find a mathematical structure that has the two properties of addition and scalar multiplication, I call the structure linear. \(\RR^n\) is a linear space because vectors allow for addition and scalar multiplication. As I said in Section 1.1, this algebraic sense of linear will be a major theme in the course. This is the first instance of that theme.

Definition 1.3.7.

The length of a vector \(u\) in \(\RR^n\) is written \(|u|\) and is given by a generalized form of the Pythagorean rule for right triangles.
\begin{equation*} |u| = \sqrt{u_1^2 + u_2^2 + \ldots + u_n^2} \end{equation*}
This length is also called the norm of the vector. A vector of length one is called a unit vector.
If I think of vectors as directions from the origin towards a point, this definition of length gives exactly what I expect: the physical length of that arrow in \(\RR^2\) and \(\RR^3\text{.}\) Past \(\RR^3\text{,}\) I don’t have a natural notion of length. This definition serves as a reasonable generalization to \(\RR^4\) and higher dimensions, which I can’t visualize. Note also that \(|u| = 0\) only if \(u\) is the zero vector. All other vectors have a positive length.
Often the square root is annoying, and I find it convenient to work with the square of the length.
\begin{equation*} |u|^2 = u_1^2 + u_2^2 + \ldots + u_n^2 \end{equation*}
The notions of length and difference allow me to define the distance between two vectors.

Definition 1.3.8.

The distance between two vectors \(u\) and \(v\) in \(\RR^n\) is the length of their difference: \(|u-v|\text{.}\)
You can check from the definition that \(|u-v| = |v-u|\text{,}\) so distance doesn’t depend on which comes first. If \(|\cdot|\) were absolute value in \(\RR\text{,}\) this definition would match the notion of distance between numbers on the number line. Difference and length are visualized in Figure 1.3.9.
Figure 1.3.9. Visualizing Distance Between Vectors
The proof of the triangle inequality is left to an exercise later in the course.