Linear Algebra › Section A · Linear equations in linear algebra › Topic A.3

Vectors and vector equations

A column of numbers can be treated as a single object that you add, scale and picture as an arrow. Seeing a linear system as one equation between vectors turns solving into a geometric question: can this target be built from these ingredients?

One object, many numbers

A list of numbers often belongs together: the three coordinates of a point, the hours a product needs in three workshops, or the red, green and blue values of a pixel. Linear algebra treats such a list as one object, a vector, and does arithmetic with it directly. This topic sets up that arithmetic and then rewrites a linear system as a single equation between vectors. That rewrite is the bridge between the row operations of the earlier topics and the geometry of lines and planes.

Vectors in the plane

A vector in \(\mathbb{R}^2\) is an ordered pair of real numbers written as a column, such as

\mathbf{u} = \begin{bmatrix} 3 \\ -1 \end{bmatrix}

To save space it is also written \((3, -1)\), but it still means a column. Two vectors are equal when their corresponding entries are equal, so \((3, -1)\) and \((-1, 3)\) differ. Vectors add entry by entry, \((3, -1) + (1, 4) = (4, 3)\), and a scalar multiple multiplies every entry, \(2(3, -1) = (6, -2)\). Here a scalar is simply a real number.

The geometric picture

Each vector \((a, b)\) can be drawn as the point \((a, b)\) or as an arrow from the origin to that point. Addition follows the parallelogram rule: if the arrows for \(\mathbf{u}\) and \(\mathbf{v}\) form two sides of a parallelogram from the origin, then \(\mathbf{u} + \mathbf{v}\) is the diagonal from the origin to the opposite corner. A scalar multiple \(c\mathbf{u}\) points the same way as \(\mathbf{u}\) when \(c > 0\), the opposite way when \(c < 0\), and has length \(|c|\) times that of \(\mathbf{u}\). All the multiples of a non-zero vector lie on one line through the origin.

Vectors in R³ and Rⁿ

A vector in \(\mathbb{R}^3\) has three entries and is pictured in space. More generally, \(\mathbb{R}^n\) is the set of all columns of \(n\) real numbers, with addition and scalar multiples defined entry by entry. The zero vector \(\mathbf{0}\) has every entry 0. For all \(\mathbf{u}\), \(\mathbf{v}\), \(\mathbf{w}\) in \(\mathbb{R}^n\) and scalars \(c\), \(d\): \(\mathbf{u} + \mathbf{v} = \mathbf{v} + \mathbf{u}\); \((\mathbf{u} + \mathbf{v}) + \mathbf{w} = \mathbf{u} + (\mathbf{v} + \mathbf{w})\); \(\mathbf{u} + \mathbf{0} = \mathbf{u}\); \(\mathbf{u} + (-\mathbf{u}) = \mathbf{0}\); \(c(\mathbf{u} + \mathbf{v}) = c\mathbf{u} + c\mathbf{v}\); \((c + d)\mathbf{u} = c\mathbf{u} + d\mathbf{u}\); \(c(d\mathbf{u}) = (cd)\mathbf{u}\); and \(1\mathbf{u} = \mathbf{u}\). Each follows from the matching rule for real numbers, applied entry by entry.

Linear combinations

Given vectors \(\mathbf{v}_1, \dots, \mathbf{v}_p\) in \(\mathbb{R}^n\) and scalars \(c_1, \dots, c_p\), the vector

\mathbf{y} = c_1\mathbf{v}_1 + \dots + c_p\mathbf{v}_p

is a linear combination of the vectors, with weights \(c_1, \dots, c_p\). The weights may be any real numbers, including zero and negatives. With \(\mathbf{u} = (1, -2)\) and \(\mathbf{v} = (3, 1)\), the combination \(3\mathbf{u} - 2\mathbf{v}\) is \((3 - 6, -6 - 2) = (-3, -8)\). In practice, suppose product A needs \((2, 1, 3)\) hours of cutting, sewing and packing, and product B needs \((1, 4, 2)\). Making 10 of A and 5 of B needs \(10(2, 1, 3) + 5(1, 4, 2)\), which is \((25, 30, 40)\) hours.

Vector equations are linear systems

Asking whether \(\mathbf{b}\) is a linear combination of \(\mathbf{a}_1, \dots, \mathbf{a}_n\) means asking whether the vector equation

x_1\mathbf{a}_1 + x_2\mathbf{a}_2 + \dots + x_n\mathbf{a}_n = \mathbf{b}

has a solution. Comparing entries turns it into an ordinary linear system, and the key fact is this: the vector equation has the same solution set as the system whose augmented matrix has the columns \(\mathbf{a}_1, \dots, \mathbf{a}_n, \mathbf{b}\) in that order. So \(\mathbf{b}\) is a linear combination of the \(\mathbf{a}_j\) exactly when that system is consistent, and row reduction decides it.

Worked example: finding the weights

Is \(\mathbf{b} = (4, 11, 5)\) a linear combination of \(\mathbf{a}_1 = (1, 2, -1)\) and \(\mathbf{a}_2 = (2, 5, 1)\)? 1. Form the augmented matrix with columns \(\mathbf{a}_1\), \(\mathbf{a}_2\), \(\mathbf{b}\): rows \((1, 2, 4)\), \((2, 5, 11)\) and \((-1, 1, 5)\). 2. Row 2 minus 2 times row 1 gives \((0, 1, 3)\); row 3 plus row 1 gives \((0, 3, 9)\). 3. Row 3 minus 3 times row 2 gives \((0, 0, 0)\), so the system is consistent. 4. Then \(x_2 = 3\) and \(x_1 = 4 - 2(3) = -2\). 5. Check: \(-2(1, 2, -1) + 3(2, 5, 1)\) is \((4, 11, 5)\). So \(\mathbf{b} = -2\mathbf{a}_1 + 3\mathbf{a}_2\).

The span of a set of vectors

The set of all linear combinations of \(\mathbf{v}_1, \dots, \mathbf{v}_p\) is called the span of those vectors, written \(\mathrm{Span}\{\mathbf{v}_1, \dots, \mathbf{v}_p\}\). It always contains the zero vector (take every weight 0) and every scalar multiple of each \(\mathbf{v}_j\). Asking whether \(\mathbf{b}\) is in the span is the same as asking whether the system with augmented matrix \([\mathbf{v}_1 \ \cdots \ \mathbf{v}_p \ \mathbf{b}]\) is consistent. Geometrically, in \(\mathbb{R}^3\) the span of one non-zero vector is a line through the origin, and the span of two vectors that are not multiples of each other is a plane through the origin.

Worked example: a vector outside the span

Is \(\mathbf{b} = (1, 1, 4)\) in \(\mathrm{Span}\{\mathbf{a}_1, \mathbf{a}_2\}\), where \(\mathbf{a}_1 = (1, 0, 2)\) and \(\mathbf{a}_2 = (0, 1, 3)\)? 1. The vector equation \(x_1\mathbf{a}_1 + x_2\mathbf{a}_2 = \mathbf{b}\) gives \(x_1 = 1\), \(x_2 = 1\) and \(2x_1 + 3x_2 = 4\). 2. The first two equations force \(x_1 = x_2 = 1\). 3. Then the third reads \(2 + 3 = 4\), which is false, so the system is inconsistent. 4. Therefore \(\mathbf{b}\) is not in the span. Geometrically, the span is a plane through the origin in \(\mathbb{R}^3\), and \(\mathbf{b}\) lies off that plane.

Common mistakes to avoid

First, putting the given vectors into the rows of the augmented matrix instead of the columns. Second, treating \((3, -1)\) and \((-1, 3)\) as the same vector: order matters. Third, forgetting that weights can be negative or zero, so missing combinations such as \(\mathbf{b} = -2\mathbf{a}_1 + 3\mathbf{a}_2\). Fourth, thinking a span can miss the origin: it always contains \(\mathbf{0}\). Fifth, assuming two vectors in \(\mathbb{R}^3\) always span a plane; if one is a multiple of the other, they span only a line, or just the origin when both are zero.

Key takeaways

Vectors in Rⁿ are columns of n real numbers that add entry by entry and scale by scalars, obeying eight familiar algebraic rules. Geometrically, addition follows the parallelogram rule and scalar multiples stretch or reverse an arrow. A linear combination is a weighted sum of vectors. The vector equation x₁a₁ + … + xₙaₙ = b has the same solutions as the system with augmented matrix [a₁ … aₙ b]. The span of a set is all its linear combinations, always contains 0, and in R³ is a line through the origin for one non-zero vector, or a plane through the origin for two vectors that are not multiples of each other.

Flashcards

Flashcard

Vector in Rⁿ

An ordered column of n real numbers

Written (a, b, c) to save space.

Test your understanding

Question 1 of 5

Let u = (4, −1, 2) and v = (1, 2, −3). What is 2u − 3v?

Question 2 of 5

Which augmented matrix matches the vector equation x₁(1, 3) + x₂(2, −1) = (5, 4)?

Question 3 of 5

In R³, what is the span of two vectors, neither of which is a multiple of the other?

Question 4 of 5

Let a₁ = (1, 1, 0), a₂ = (0, 1, 1) and b = (2, 5, 3). Is b a linear combination of a₁ and a₂?

Question 5 of 5

Which vector is always in Span{v₁, v₂}, whatever v₁ and v₂ in R³ are?

← Previous topicNext topic →