Linear Systems, Matrix Algebra, and Analytic Geometry

Systems of Linear Equations

  • Definition of a Linear Equation:

    • A linear equation in nn variables has the general standard form:     a1x1+a2x2+⋯+anxn=ba_1 x_1 + a_2 x_2 + \dots + a_n x_n = b

    • The numbers a1,a2,…,ana_1, a_2, \dots, a_n are constants called the coefficients of the variables, and bb is a constant called the constant term.

    • For two variables, this corresponds to the line in R2\mathbb{R}^2: ax+by=cax + by = c.

    • For three variables, this corresponds to the plane in R3\mathbb{R}^3: ax+by+cz=dax + by + cz = d.

  • Classifying Equations as Linear or Non-Linear:

    • 3x−4y=−13x - 4y = -1: Linear. All variables are to the first power with constant coefficients.

    • xy+2z=1xy + 2z = 1: Non-linear. Contains a product of variables (xyxy).

    • xy+z=2\frac{x}{y} + z = 2: Non-linear. Involves division by a variable (xy\frac{x}{y}).

    • r−12s−5t=9r - \frac{1}{2}s - 5t = 9: Linear. Constant coefficients with variables to the first power.

    • x12−x32=3x_1^2 - x_3^2 = 3: Non-linear. Variables are squared (x12,x32x_1^2, x_3^2).

    • 3.2x1−0.01x2=4.63.2x_1 - 0.01x_2 = 4.6: Linear. Real decimal coefficients, degree-one variables.

    • x1+5x2=3−x3+2x4x_1 + 5x_2 = 3 - x_3 + 2x4: Linear. Rearranges to x1+5x2+x3−2x4=3x_1 + 5x_2 + x_3 - 2x_4 = 3.

    • 2x+π4y−sin⁡(π/5)z=1\sqrt{2}x + \frac{\pi}{4}y - \sin(\pi/5)z = 1: Linear. The coefficient sin⁡(π/5)\sin(\pi/5) is a constant scalar multiplying zz.

    • 2x+π4y−sin⁡((π/5)z)=1\sqrt{2}x + \frac{\pi}{4}y - \sin((\pi/5)z) = 1: Non-linear. The variable zz is inside a non-linear trigonometric function sin⁡\sin

  • Solutions to Linear Equations:

    • A column vector s=[s1s2…sn]T\mathbf{s} = \begin{bmatrix} s_1 & s_2 & \dots & s_n \end{bmatrix}^T is a solution to a1x1+⋯+anxn=ba_1 x_1 + \dots + a_n x_n = b if substitution yields the true scalar identity:     a1s1+a2s2+⋯+ansn=ba_1 s_1 + a_2 s_2 + \dots + a_n s_n = b

    • For 3x−4y=−13x - 4y = -1:

    • Setting x=1x = 1 gives 3(1)−4y=−1  ⟹  −4y=−4  ⟹  y=13(1) - 4y = -1 \implies -4y = -4 \implies y = 1, yielding solution [11]\begin{bmatrix} 1 \\ 1 \end{bmatrix}.

    • Setting x=−3x = -3 gives 3(−3)−4y=−1  ⟹  −4y=8  ⟹  y=−23(-3) - 4y = -1 \implies -4y = 8 \implies y = -2, yielding solution [−3−2]\begin{bmatrix} -3 \\ -2 \end{bmatrix}.

    • Geometrically, the full solution set represents a straight line in R2\mathbb{R}^2.

  • Systems of Linear Equations:

    • A system of linear equations is a finite set of linear equations involving the same set of variables.

    • A solution to a system is a vector that simultaneously satisfies every equation in the system.

    • Two systems are defined as equivalent if and only if they possess the exact same solution set.

  • Fundamental Outcome Theorem for Real Linear Systems:

    • Every real linear system has exactly one of the following three outcomes:

    1. Exactly one unique solution.

    2. Infinitely many solutions.

    3. No solution.

    • A system is termed consistent if it possesses at least one solution (outcomes 1 or 2).

    • A system is termed inconsistent if it possesses no solution (outcome 3).

  • Geometric Examples of System Outcomes:

    • Unique Solution:     {x−y=1x+y=3\begin{cases} x - y = 1 \\ x + y = 3 \end{cases}     Adding both equations yields 2x=4  ⟹  x=22x = 4 \implies x = 2, so y=1y = 1. The solution is the single point [21]\begin{bmatrix} 2 \\ 1 \end{bmatrix}. Geometrically, this represents two non-parallel lines intersecting at a single point.

    • Infinitely Many Solutions:     {x−y=22x−2y=4\begin{cases} x - y = 2 \\ 2x - 2y = 4 \end{cases}     The second equation is a multiple of the first (2×2 \times first equation). Any point on x−y=2x - y = 2 satisfies both equations. Geometrically, these equations represent coincident lines.

    • No Solution:     {x−y=1x−y=3\begin{cases} x - y = 1 \\ x - y = 3 \end{cases}     Subtracting the equations yields 0=−20 = -2, a contradiction. Geometrically, these equations represent parallel and distinct lines in R2\mathbb{R}^2.

  • Solving Systems by Back Substitution:

    • A system in triangular form can be solved directly from bottom to top.

    • Consider the system:     {x−y−z=2y+3z=55z=10\begin{cases} x - y - z = 2 \\ y + 3z = 5 \\ 5z = 10 \end{cases}

    • Solve the bottom equation: 5z=10  ⟹  z=25z = 10 \implies z = 2.

    • Substitute z=2z = 2 into the middle equation: y+3(2)=5  ⟹  y=−1y + 3(2) = 5 \implies y = -1.

    • Substitute y=−1y = -1 and z=2z = 2 into the top equation: x−(−1)−2=2  ⟹  x−1=2  ⟹  x=3x - (-1) - 2 = 2 \implies x - 1 = 2 \implies x = 3.

    • The unique solution vector is x=[3−12]T\mathbf{x} = \begin{bmatrix} 3 & -1 & 2 \end{bmatrix}^T.

  • Matrix Representation of Systems:

    • For a system Ax=bA\mathbf{x} = \mathbf{b}:

    • The coefficient matrix AA contains exclusively the coefficients of the variables.

    • The augmented matrix [A∣b][A \mid \mathbf{b}] contains the coefficient columns augmented by the constant column b\mathbf{b}.

    • Example conversion:     {2x+y−z=3x+5z=1−x+3y−2z=0\begin{cases} 2x + y - z = 3 \\ x + 5z = 1 \\ -x + 3y - 2z = 0 \end{cases}

    • Coefficient matrix:       A=[21−1105−13−2]A = \begin{bmatrix} 2 & 1 & -1 \\ 1 & 0 & 5 \\ -1 & 3 & -2 \end{bmatrix}

    • Augmented matrix:       [A∣b]=[21−131051−13−20][A \mid \mathbf{b}] = \left[\begin{array}{ccc|c} 2 & 1 & -1 & 3 \\ 1 & 0 & 5 & 1 \\ -1 & 3 & -2 & 0 \end{array}\right]

  • Row Echelon Form (REF):

    • A matrix is in Row Echelon Form (REF) if:

    1. All all-zero rows are grouped together at the bottom of the matrix.

    2. In each non-zero row, the first non-zero entry (the leading entry or pivot) lies strictly to the right of the leading entry of the row directly above it.

    • This creates a characteristic staircase pattern.

    • Valid REF matrices:     [2410−12000],[101015004],[112100130000]\begin{bmatrix} 2 & 4 & 1 \\ 0 & -1 & 2 \\ 0 & 0 & 0 \end{bmatrix}, \quad \begin{bmatrix} 1 & 0 & 1 \\ 0 & 1 & 5 \\ 0 & 0 & 4 \end{bmatrix}, \quad \left[\begin{array}{ccc|c} 1 & 1 & 2 & 1 \\ 0 & 0 & 1 & 3 \\ 0 & 0 & 0 & 0 \end{array}\right]

  • Elementary Row Operations:

    • Elementary row operations transform an augmented matrix into an equivalent augmented matrix without changing the solution set:

    1. Interchange two rows: Ri↔RjR_i \leftrightarrow R_j

    2. Multiply a row by a non-zero scalar: Ri←kRi(k≠0)R_i \leftarrow k R_i \quad (k \neq 0)

    3. Add a scalar multiple of one row to another row: Ri←Ri+kRjR_i \leftarrow R_i + k R_j

  • Gaussian Elimination and Gauss-Jordan Elimination:

    • Gaussian Elimination: Transform the augmented matrix to REF using elementary row operations, then extract the solution using back substitution.

    • Reduced Row Echelon Form (RREF): A matrix is in RREF if:

    1. It is in REF.

    2. Every leading entry in a non-zero row is equal to 11 (a leading 1).

    3. Every column containing a leading 1 has zeros in all other positions (both above and below).

    • Gauss-Jordan Elimination: Transform the augmented matrix to REF, verify consistency, and continue row operations upward until the matrix reaches RREF.

  • Worked Row Reduction (Example 5 & Example 5 Redux):

    • Consider the augmented matrix:     [1−1−1212−2−133−11−10−3]\left[\begin{array}{cccc|c} 1 & -1 & -1 & 2 & 1 \\ 2 & -2 & -1 & 3 & 3 \\ -1 & 1 & -1 & 0 & -3 \end{array}\right]

    • Elimination to REF:

    • Perform R2←R2−2R1R_2 \leftarrow R_2 - 2R_1 and R3←R3+R1R_3 \leftarrow R_3 + R_1:       [1−1−121001−1100−22−2]\left[\begin{array}{cccc|c} 1 & -1 & -1 & 2 & 1 \\ 0 & 0 & 1 & -1 & 1 \\ 0 & 0 & -2 & 2 & -2 \end{array}\right]

    • Perform R3←R3+2R2R_3 \leftarrow R_3 + 2R_2:       [1−1−121001−1100000]\left[\begin{array}{cccc|c} 1 & -1 & -1 & 2 & 1 \\ 0 & 0 & 1 & -1 & 1 \\ 0 & 0 & 0 & 0 & 0 \end{array}\right]

    • The matrix is now in REF.

    • Continuation to RREF:

    • Clear above the leading 1 in row 2: perform R1←R1+R2R_1 \leftarrow R_1 + R_2:       [1−1012001−1100000]\left[\begin{array}{cccc|c} 1 & -1 & 0 & 1 & 2 \\ 0 & 0 & 1 & -1 & 1 \\ 0 & 0 & 0 & 0 & 0 \end{array}\right]

    • Parametric Solution:

    • Leading variables (corresponding to columns with leading 1s): x1,x3x_1, x_3.

    • Free variables: x2=s,x4=tx_2 = s, x_4 = t.

    • Row equations: x1−s+t=2  ⟹  x1=2+s−tx_1 - s + t = 2 \implies x_1 = 2 + s - t; x3−t=1  ⟹  x3=1+tx_3 - t = 1 \implies x_3 = 1 + t.

    • Vector form:       x=[x1x2x3x4]=[2+s−ts1+tt]=[2010]+s[1100]+t[−1011],s,t∈R\mathbf{x} = \begin{bmatrix} x_1 \\ x_2 \\ x_3 \\ x_4 \end{bmatrix} = \begin{bmatrix} 2 + s - t \\ s \\ 1 + t \\ t \end{bmatrix} = \begin{bmatrix} 2 \\ 0 \\ 1 \\ 0 \end{bmatrix} + s \begin{bmatrix} 1 \\ 1 \\ 0 \\ 0 \end{bmatrix} + t \begin{bmatrix} -1 \\ 0 \\ 1 \\ 1 \end{bmatrix}, \quad s, t \in \mathbb{R}

  • Rank and the Rank Theorem:

    • Definition of Rank: The rank of a matrix AA, denoted rank(A)\text{rank}(A), is the total number of leading entries (pivots) present after AA is reduced to Row Echelon Form.

    • Rank Theorem (Version 1): For a consistent linear system having coefficient matrix AA with nn variables:     number of parameters=n−rank(A)\text{number of parameters} = n - \text{rank}(A)

    • Parameter and Shape Analysis Table:

    • System 1:       [1−1−2−501130012]\left[\begin{array}{ccc|c} 1 & -1 & -2 & -5 \\ 0 & 1 & 1 & 3 \\ 0 & 0 & 1 & 2 \end{array}\right]       Variables: 33. Leading entries: 33. Parameters: 3−3=03 - 3 = 0. Solutions: 11 unique solution. Shape: Single point.

    • System 2:       [112100130000]\left[\begin{array}{ccc|c} 1 & 1 & 2 & 1 \\ 0 & 0 & 1 & 3 \\ 0 & 0 & 0 & 0 \end{array}\right]       Variables: 33. Leading entries: 22 (x1,x3x_1, x_3). Parameters: 3−2=13 - 2 = 1. Solutions: Infinitely many. Shape: Line in R3\mathbb{R}^3.

    • System 3:       [1−1012001−1100000]\left[\begin{array}{cccc|c} 1 & -1 & 0 & 1 & 2 \\ 0 & 0 & 1 & -1 & 1 \\ 0 & 0 & 0 & 0 & 0 \end{array}\right]       Variables: 44. Leading entries: 22 (x1,x3x_1, x_3). Parameters: 4−2=24 - 2 = 2. Solutions: Infinitely many. Shape: Two-dimensional plane in R4\mathbb{R}^4.

    • Example Analysis without Explicit Solving:     [105−701−35]\left[\begin{array}{ccc|c} 1 & 0 & 5 & -7 \\ 0 & 1 & -3 & 5 \end{array}\right]     Number of variables: n=3n = 3. Rank of coefficient matrix: rank(A)=2\text{rank}(A) = 2. Since it is consistent, number of parameters is 3−2=13 - 2 = 1. The geometric shape of the solution set is a straight line in R3\mathbb{R}^3.

  • Homogeneous Systems:

    • A linear system is defined as homogeneous if all constant terms are zero:     Ax=0or[A∣0]A\mathbf{x} = \mathbf{0} \quad \text{or} \quad [A \mid \mathbf{0}]

    • Every homogeneous system is guaranteed to have at least the trivial solution:     x=0=[00…0]T\mathbf{x} = \mathbf{0} = \begin{bmatrix} 0 & 0 & \dots & 0 \end{bmatrix}^T

    • A homogeneous system can never be inconsistent (it always has at least one solution).

    • Underdetermined Homogeneous Systems Theorem: If a homogeneous system has mm equations and nn variables with fewer equations than variables (m<nm < n), the system is guaranteed to possess infinitely many solutions.

    • Justification: rank(A)≤m<n\text{rank}(A) \le m < n. Therefore, the number of free parameters n−rank(A)≥n−m>0n - \text{rank}(A) \ge n - m > 0. At least one free parameter exists.

Matrix Algebra and Special Matrices

  • Definition of a Matrix:

    • An m×nm \times n matrix is a rectangular array of numbers arranged into mm horizontal rows and nn vertical columns:     A=[aij]=[a11a12…a1na21a22…a2n⋮⋮⋱⋮am1am2…amn]A = [a_{ij}] = \begin{bmatrix} a_{11} & a_{12} & \dots & a_{1n} \\ a_{21} & a_{22} & \dots & a_{2n} \\ \vdots & \vdots & \ddots & \vdots \\ a_{m1} & a_{m2} & \dots & a_{mn} \end{bmatrix}

    • The scalar aija_{ij} represents the entry located in row ii and column jj.

    • A matrix can also be expressed compactly by writing its column vectors:     A=[a⃗1a⃗2…a⃗n]A = \begin{bmatrix} \vec{a}_1 & \vec{a}_2 & \dots & \vec{a}_n \end{bmatrix}

    • The entries a11,a22,a33,…a_{11}, a_{22}, a_{33}, \dots form the main diagonal of the matrix (where the row index equals the column index).

  • Notation Check:

    • For the matrix:     A=[123456]A = \begin{bmatrix} 1 & 2 & 3 \\ 4 & 5 & 6 \end{bmatrix}

    • Size: 2×32 \times 3 (22 rows, 33 columns).

    • Entry a12=2a_{12} = 2 (row 1, column 2).

    • Entry a21=4a_{21} = 4 (row 2, column 1).

    • Second column: [25]\begin{bmatrix} 2 \\ 5 \end{bmatrix}.

  • Special Matrix Classes:

    • Zero Matrix (OO): Every entry equals 00. Its dimensions depend on context (e.g., O2×3=[000000]O_{2 \times 3} = \begin{bmatrix} 0 & 0 & 0 \\ 0 & 0 & 0 \end{bmatrix}).

    • Diagonal Matrix: A square matrix where all entries outside the main diagonal are zero (aij=0a_{ij} = 0 for all i≠ji \neq j), such as [2005]\begin{bmatrix} 2 & 0 \\ 0 & 5 \end{bmatrix}.

    • Scalar Matrix: A diagonal matrix whose main diagonal entries are all identical, equal to cInc I_n (e.g., 3I2=[3003]3I_2 = \begin{bmatrix} 3 & 0 \\ 0 & 3 \end{bmatrix}).

    • Identity Matrix (InI_n): A square n×nn \times n matrix with 11s along the main diagonal and 00s everywhere else:     I3=[100010001]I_3 = \begin{bmatrix} 1 & 0 & 0 \\ 0 & 1 & 0 \\ 0 & 0 & 1 \end{bmatrix}

  • Matrix Addition and Scalar Multiplication:

    • Matrices must have identical dimensions to be added or subtracted.

    • If A=[aij]A = [a_{ij}] and B=[bij]B = [b_{ij}] are both m×nm \times n, then:     A+B=[aij+bij]A + B = [a_{ij} + b_{ij}]     cA=[caij](c∈R)cA = [c a_{ij}] \quad (c \in \mathbb{R})

    • Subtraction is defined by A−B=A+(−1)B=[aij−bij]A - B = A + (-1)B = [a_{ij} - b_{ij}].

  • Algebraic Rules for Addition and Scaling:

    • For same-size matrices A,B,CA, B, C and real scalars c,dc, d:

    1. Commutativity of Addition: A+B=B+AA + B = B + A

    2. Additive Identity: A+O=AA + O = A

    3. Additive Inverse: A+(−A)=OA + (-A) = O

    4. Associativity of Addition: A+(B+C)=(A+B)+CA + (B + C) = (A + B) + C

    5. Distributive over Matrix Addition: c(A+B)=cA+cBc(A + B) = cA + cB

    6. Distributive over Scalar Addition: (c+d)A=cA+dA(c + d)A = cA + dA

    7. Scalar Associativity: c(dA)=(cd)Ac(dA) = (cd)A

    8. Identity Scaling: 1A=A1A = A

  • Matrix Algebraic Computations:

    • Combining Like Terms:     3A+(5B−2A)+2(B−A)=3A+5B−2A+2B−2A=(3−2−2)A+(5+2)B=−A+7B3A + (5B - 2A) + 2(B - A) = 3A + 5B - 2A + 2B - 2A = (3 - 2 - 2)A + (5 + 2)B = -A + 7B

    • Solving Linear Matrix Equations:     5X−A=2(A+2X)  ⟹  5X−A=2A+4X  ⟹  5X−4X=2A+A  ⟹  X=3A5X - A = 2(A + 2X) \implies 5X - A = 2A + 4X \implies 5X - 4X = 2A + A \implies X = 3A

  • Linear Combinations and Span of Matrices:

    • If A1,A2,…,AkA_1, A_2, \dots, A_k have the same dimensions, a linear combination is:     c1A1+c2A2+⋯+ckAk(c1,…,ck∈R)c_1 A_1 + c_2 A_2 + \dots + c_k A_k \quad (c_1, \dots, c_k \in \mathbb{R})

    • The span of a set of matrices is the set of all possible linear combinations that can be formed from them.

    • Given:     A1=[01−10],A2=[1001],A3=[1111]A_1 = \begin{bmatrix} 0 & 1 \\ -1 & 0 \end{bmatrix}, \quad A_2 = \begin{bmatrix} 1 & 0 \\ 0 & 1 \end{bmatrix}, \quad A_3 = \begin{bmatrix} 1 & 1 \\ 1 & 1 \end{bmatrix}

    • General combination:       xA1+yA2+zA3=x[01−10]+y[1001]+z[1111]=[y+zx+z−x+zy+z]x A_1 + y A_2 + z A_3 = x \begin{bmatrix} 0 & 1 \\ -1 & 0 \end{bmatrix} + y \begin{bmatrix} 1 & 0 \\ 0 & 1 \end{bmatrix} + z \begin{bmatrix} 1 & 1 \\ 1 & 1 \end{bmatrix} = \begin{bmatrix} y + z & x + z \\ -x + z & y + z \end{bmatrix}

    • Crucial structural property: The two diagonal entries are strictly equal (a11=a22=y+za_{11} = a_{22} = y + z).

    • Testing B=[1421]B = \begin{bmatrix} 1 & 4 \\ 2 & 1 \end{bmatrix}:

      • Diagonal entries match (1=11 = 1), so y+z=1y + z = 1.

      • Off-diagonal: x+z=4x + z = 4 and −x+z=2-x + z = 2.

      • Adding off-diagonal equations gives 2z=6  ⟹  z=32z = 6 \implies z = 3.

      • Then x=4−3=1x = 4 - 3 = 1, and y=1−3=−2y = 1 - 3 = -2.

      • Since consistent, B=A1−2A2+3A3B = A_1 - 2A_2 + 3A_3. Thus, B∈span{A1,A2,A3}B \in \text{span}\{A_1, A_2, A_3\}.

    • Testing C=[1234]C = \begin{bmatrix} 1 & 2 \\ 3 & 4 \end{bmatrix}:

      • Top-left entry requires y+z=1y + z = 1.

      • Bottom-right entry requires y+z=4y + z = 4.

      • Since a scalar cannot simultaneously equal 11 and 44, this system is inconsistent. Thus, C∉span{A1,A2,A3}C \notin \text{span}\{A_1, A_2, A_3\}.

  • Description of Full Span:   span{A1,A2,A3}={[abca]:a,b,c∈R}\text{span}\{A_1, A_2, A_3\} = \left\{ \begin{bmatrix} a & b \\ c & a \end{bmatrix} : a, b, c \in \mathbb{R} \right\}   For any real values a,b,ca, b, c, the unique combination coefficients are:   z=b+c2,x=b−c2,y=a−b+c2z = \frac{b + c}{2}, \quad x = \frac{b - c}{2}, \quad y = a - \frac{b + c}{2}

  • Linear Independence of the Matrices:

    • Setting the linear combination to the zero matrix:     [y+zx+z−x+zy+z]=[0000]\begin{bmatrix} y + z & x + z \\ -x + z & y + z \end{bmatrix} = \begin{bmatrix} 0 & 0 \\ 0 & 0 \end{bmatrix}

    • System: x+z=0x + z = 0, −x+z=0-x + z = 0, y+z=0y + z = 0.

    • Adding the first two equations gives 2z=0  ⟹  z=02z = 0 \implies z = 0, which forces x=0x = 0 and y=0y = 0.

    • Only the trivial solution exists; therefore, the set of matrices {A1,A2,A3}\{A_1, A_2, A_3\} is linearly independent.

Matrix Operations and Transpose Identities

  • Definition of Matrix Product:

    • Let AA be an m×km \times k matrix and BB be a k×rk \times r matrix. The product ABAB is an m×rm \times r matrix whose (i,j)(i, j) entry is given by:     (AB)ij=∑ℓ=1kaiℓbℓj=ai1b1j+ai2b2j+⋯+aikbkj(AB)_{ij} = \sum_{\ell=1}^k a_{i\ell} b_{\ell j} = a_{i1} b_{1j} + a_{i2} b_{2j} + \dots + a_{ik} b_{kj}

    • Entry (i,j)(i, j) is the dot product of row ii of matrix AA with column jj of matrix BB.

    • Multiplication requirement: The inner dimensions must match (k=kk = k). The outer dimensions (m×rm \times r) dictate the dimensions of the product matrix:     (m×k)×(k×r)⟶(m×r)(m \times k) \times (k \times r) \longrightarrow (m \times r)

  • Properties of Matrix Multiplication:

    • For dimensionally compatible matrices and scalar c∈Rc \in \mathbb{R}:

    1. Associative Law: A(BC)=(AB)CA(BC) = (AB)C

    2. Left Distributive Law: A(B+C)=AB+ACA(B + C) = AB + AC

    3. Right Distributive Law: (A+B)C=AC+BC(A + B)C = AC + BC

    4. Scalar Compatibility: c(AB)=(cA)B=A(cB)c(AB) = (cA)B = A(cB)

    5. Identity Multiplicative Property: For an m×nm \times n matrix AA, ImA=A=AInI_m A = A = A I_n.

  • Failures of Real Number Arithmetic in Matrix Multiplication:

    • Non-Commutativity: In general, AB≠BAAB \neq BA, even when both products are defined and have the same dimensions.

  • Zero Divisors (Cancellation Failure): For real numbers, ab=0  ⟹  a=0 or b=0ab = 0 \implies a = 0 \text{ or } b = 0. For matrices, AB=OAB = O does not imply A=OA = O or B=OB = O.

    • Explicit counterexample:     A=[0100],B=[1000]A = \begin{bmatrix} 0 & 1 \\ 0 & 0 \end{bmatrix}, \quad B = \begin{bmatrix} 1 & 0 \\ 0 & 0 \end{bmatrix}     AB=[0100][1000]=[0000]=OAB = \begin{bmatrix} 0 & 1 \\ 0 & 0 \end{bmatrix} \begin{bmatrix} 1 & 0 \\ 0 & 0 \end{bmatrix} = \begin{bmatrix} 0 & 0 \\ 0 & 0 \end{bmatrix} = O     BA=[1000][0100]=[0100]≠OBA = \begin{bmatrix} 1 & 0 \\ 0 & 0 \end{bmatrix} \begin{bmatrix} 0 & 1 \\ 0 & 0 \end{bmatrix} = \begin{bmatrix} 0 & 1 \\ 0 & 0 \end{bmatrix} \neq O       Both AA and BB are non-zero, yet AB=OAB = O. Furthermore, AB≠BAAB \neq BA.

  • Matrix Powers:

    • Matrix powers are defined only for square matrices (n×nn \times n). For integer k≥0k \ge 0:     A0=In,A1=A,Ak=AA…A⏟k factors(k≥1)A^0 = I_n, \quad A^1 = A, \quad A^k = \underbrace{A A \dots A}_{k \text{ factors}} \quad (k \ge 1)

    • Exponent rule: ArAs=Ar+sA^r A^s = A^{r+s} for non-negative integers r,sr, s.

    • Example Cycle of Powers:

    • Let B=[0−110]B = \begin{bmatrix} 0 & -1 \\ 1 & 0 \end{bmatrix}.

    • B2=[0−110][0−110]=[−100−1]=−I2B^2 = \begin{bmatrix} 0 & -1 \\ 1 & 0 \end{bmatrix} \begin{bmatrix} 0 & -1 \\ 1 & 0 \end{bmatrix} = \begin{bmatrix} -1 & 0 \\ 0 & -1 \end{bmatrix} = -I_2

    • B3=B2B=(−I2)B=−B=[01−10]B^3 = B^2 B = (-I_2)B = -B = \begin{bmatrix} 0 & 1 \\ -1 & 0 \end{bmatrix}

    • B4=(B2)2=(−I2)2=I2=[1001]B^4 = (B^2)^2 = (-I_2)^2 = I_2 = \begin{bmatrix} 1 & 0 \\ 0 & 1 \end{bmatrix}

    • The powers cycle with period 4: B5=B,B6=−I2,B7=−B,B8=I2B^5 = B, B^6 = -I_2, B^7 = -B, B^8 = I_2.

  • Expansion of Binomial Products:

    • For square matrices AA and BB of the same size:     (A+B)2=(A+B)(A+B)=A2+AB+BA+B2(A + B)^2 = (A + B)(A + B) = A^2 + AB + BA + B^2

    • The familiar elementary algebraic formula (A+B)2=A2+2AB+B2(A + B)^2 = A^2 + 2AB + B^2 holds if and only if AB=BAAB = BA (that is, AA and BB commute).

  • Block Matrix Multiplication:

    • Large matrices can be partitioned into submatrices (blocks).

    • When the column partitions of the first matrix match the row partitions of the second matrix, block multiplication proceeds identically to regular matrix multiplication, treating submatrices as elements.

    • Zero and identity blocks within partitions dramatically accelerate computations.

  • Matrix Transpose:

    • For an m×nm \times n matrix A=[aij]A = [a_{ij}], the transpose ATA^T is the n×mn \times m matrix produced by interchanging the rows and columns:     (AT)ij=aji(A^T)_{ij} = a_{ji}

    • An entry at row ii, column jj transitions to row jj, column ii.

    • Examples:

    • A=[132501]  ⟹  AT=[153021]A = \begin{bmatrix} 1 & 3 & 2 \\ 5 & 0 & 1 \end{bmatrix} \implies A^T = \begin{bmatrix} 1 & 5 \\ 3 & 0 \\ 2 & 1 \end{bmatrix}

    • B=[abcd]  ⟹  BT=[acbd]B = \begin{bmatrix} a & b \\ c & d \end{bmatrix} \implies B^T = \begin{bmatrix} a & c \\ b & d \end{bmatrix}

    • C=[5−124]  ⟹  CT=[5−124]C = \begin{bmatrix} 5 \\ -1 \\ 2 \\ 4 \end{bmatrix} \implies C^T = \begin{bmatrix} 5 & -1 & 2 & 4 \end{bmatrix}

    • For D=[2−10473]D = \begin{bmatrix} 2 & -1 \\ 0 & 4 \\ 7 & 3 \end{bmatrix} of size 3×23 \times 2, its transpose DT=[207−143]D^T = \begin{bmatrix} 2 & 0 & 7 \\ -1 & 4 & 3 \end{bmatrix} is size 2×32 \times 3. Entry 77 at position (3,1)(3, 1) relocates to position (1,3)(1, 3).

  • Transpose Algebraic Identities:

    • For compatible matrices A,BA, B, scalar c∈Rc \in \mathbb{R}, and integer r≥0r \ge 0:

    1. Transpose of a Transpose: (AT)T=A(A^T)^T = A

    2. Transpose of a Sum: (A+B)T=AT+BT(A + B)^T = A^T + B^T

    3. Transpose of a Scalar Multiple: (cA)T=cAT(cA)^T = c A^T

    4. Transpose of a Product (Order Reversal): (AB)T=BTAT(AB)^T = B^T A^T

    5. Transpose of a Power: (Ar)T=(AT)r(A^r)^T = (A^T)^r

    • Transpose of a Triple Product: (ABC)T=CTBTAT(ABC)^T = C^T B^T A^T.

  • Entry-Level Proof of the Product Transpose Rule (AB)T=BTAT(AB)^T = B^T A^T:

    • Let AA be m×km \times k and BB be k×rk \times r. Then ABAB is m×rm \times r, and (AB)T(AB)^T is r×mr \times m.

    • The (i,j)(i, j) entry of (AB)T(AB)^T is the (j,i)(j, i) entry of ABAB:     [(AB)T]ij=(AB)ji=∑ℓ=1kajℓbℓi[(AB)^T]_{ij} = (AB)_{ji} = \sum_{\ell=1}^k a_{j\ell} b_{\ell i}

    • Now consider the matrix product BTATB^T A^T. Matrix BTB^T is r×kr \times k and ATA^T is k×mk \times m, so their product is r×mr \times m.

    • The (i,j)(i, j) entry of BTATB^T A^T is:     [BTAT]ij=∑ℓ=1k(BT)iℓ(AT)ℓj=∑ℓ=1kbℓiajℓ=∑ℓ=1kajℓbℓi[B^T A^T]_{ij} = \sum_{\ell=1}^k (B^T)_{i\ell} (A^T)_{\ell j} = \sum_{\ell=1}^k b_{\ell i} a_{j\ell} = \sum_{\ell=1}^k a_{j\ell} b_{\ell i}

    • Because real number multiplication is commutative (bℓiajℓ=ajℓbℓib_{\ell i} a_{j\ell} = a_{j\ell} b_{\ell i}), the entries are identical for all i,ji, j. Thus, (AB)T=BTAT(AB)^T = B^T A^T.

  • Dot Product and Vector Norm via Transpose:

    • For column vectors u⃗,v⃗∈Rn\vec{u}, \vec{v} \in \mathbb{R}^n:

    • Dot Product: u⃗⋅v⃗=u⃗Tv⃗\vec{u} \cdot \vec{v} = \vec{u}^T \vec{v}

      • Dimensions: (1×n)(n×1)→(1×1)(1 \times n)(n \times 1) \to (1 \times 1).

      • Example: For u⃗=[123]\vec{u} = \begin{bmatrix} 1 \\ 2 \\ 3 \end{bmatrix} and v⃗=[4−12]\vec{v} = \begin{bmatrix} 4 \\ -1 \\ 2 \end{bmatrix}:         u⃗Tv⃗=1(4)+2(−1)+3(2)=4−2+6=8\vec{u}^T \vec{v} = 1(4) + 2(-1) + 3(2) = 4 - 2 + 6 = 8

    • Squared Norm: ∥v⃗∥2=v⃗Tv⃗=v12+v22+⋯+vn2\lVert\vec{v}\rVert^2 = \vec{v}^T \vec{v} = v_1^2 + v_2^2 + \dots + v_n^2

    • Norm (Length): ∥v⃗∥=v⃗Tv⃗\lVert\vec{v}\rVert = \sqrt{\vec{v}^T \vec{v}}

      • Example: For v⃗=[34]T\vec{v} = \begin{bmatrix} 3 & 4 \end{bmatrix}^T, v⃗Tv⃗=9+16=25  ⟹  ∥v⃗∥=5\vec{v}^T \vec{v} = 9 + 16 = 25 \implies \lVert\vec{v}\rVert = 5.

  • Extracting Rows and Columns with Standard Basis Vectors:

    • Let e⃗j\vec{e}_j denote the jj-th standard basis vector (having 11 in entry jj and 00 elsewhere):

    1. Row Selection: e⃗jTA\vec{e}_j^T A selects row jj of AA (where e⃗j∈Rm\vec{e}_j \in \mathbb{R}^m).

      • In R2\mathbb{R}^2, with e⃗2=[01]\vec{e}_2 = \begin{bmatrix} 0 \\ 1 \end{bmatrix}:          e⃗2TA=[01][132501]=0[row 1]+1[row 2]=[501]\vec{e}_2^T A = \begin{bmatrix} 0 & 1 \end{bmatrix} \begin{bmatrix} 1 & 3 & 2 \\ 5 & 0 & 1 \end{bmatrix} = 0[\text{row 1}] + 1[\text{row 2}] = \begin{bmatrix} 5 & 0 & 1 \end{bmatrix}

    2. Column Selection: Ae⃗jA \vec{e}_j selects column jj of AA (where e⃗j∈Rn\vec{e}_j \in \mathbb{R}^n).

      • In R3\mathbb{R}^3, with e⃗2=[010]\vec{e}_2 = \begin{bmatrix} 0 \\ 1 \\ 0 \end{bmatrix}:          Ae⃗2=[132501][010]=0[col 1]+1[col 2]+0[col 3]=[30]A \vec{e}_2 = \begin{bmatrix} 1 & 3 & 2 \\ 5 & 0 & 1 \end{bmatrix} \begin{bmatrix} 0 \\ 1 \\ 0 \end{bmatrix} = 0[\text{col 1}] + 1[\text{col 2}] + 0[\text{col 3}] = \begin{bmatrix} 3 \\ 0 \end{bmatrix}

Vector Equations of Lines and Planes

  • Lines in Rn\mathbb{R}^n:

    • A point PP and a non-zero direction vector u⃗\vec{u} determine a unique line ℓ\ell:     ℓ:x⃗=p⃗+su⃗,s∈R,p⃗=OP⃗\ell: \vec{x} = \vec{p} + s \vec{u}, \quad s \in \mathbb{R}, \quad \vec{p} = \vec{OP}

    • Interpretation: The vector p⃗\vec{p} locates the starting point PP, the vector u⃗\vec{u} defines the line's orientation in space, and the parameter ss dictates the scalar multiple of u⃗\vec{u} added to p⃗\vec{p}.

    • Worked Example:     x⃗=[−150]+s[3−41]\vec{x} = \begin{bmatrix} -1 \\ 5 \\ 0 \end{bmatrix} + s \begin{bmatrix} 3 \\ -4 \\ 1 \end{bmatrix}

    • Parametric Equations:       {x=−1+3sy=5−4sz=s\begin{cases} x = -1 + 3s \\ y = 5 - 4s \\ z = s \end{cases}

    • At s=0s = 0, the point is P=(−1,5,0)P = (-1, 5, 0).

    • At s=1s = 1, the point reached is Q=(2,1,1)Q = (2, 1, 1).

    • For any point to lie on the line, all coordinates must be produced simultaneously by the exact same parameter value ss.

    • Multiple Descriptions of the Same Line:

    • A line possesses infinitely many valid vector equations.

    • Any point lying on the line can serve as the base point p⃗\vec{p}.

    • The direction vector u⃗\vec{u} may be replaced by any non-zero scalar multiple cu⃗c\vec{u} (c≠0c \neq 0).

    • A negative scalar multiple (c<0c < 0) reverses the directional orientation of traversal along the line, but leaves the physical set of points comprising the line unchanged.

  • Planes in Rn\mathbb{R}^n:

    • For a point P∈RnP \in \mathbb{R}^n and two non-zero, non-parallel vectors u,v∈Rn\mathbf{u}, \mathbf{v} \in \mathbb{R}^n, a two-dimensional plane is defined by:     x=p+su+tv,s,t∈R\mathbf{x} = \mathbf{p} + s\mathbf{u} + t\mathbf{v}, \quad s, t \in \mathbb{R}

    • This parametric definition describes a two-dimensional flat surface regardless of the ambient dimension nn of the space.

    • Worked Example in R4\mathbb{R}^4 (Example 18):

    • Point P=(1,0,−2,3)P = (1, 0, -2, 3), directions u=[3−121]\mathbf{u} = \begin{bmatrix} 3 \\ -1 \\ 2 \\ 1 \end{bmatrix}, v=[0112]\mathbf{v} = \begin{bmatrix} 0 \\ 1 \\ 1 \\ 2 \end{bmatrix}.

    • Vector Equation:       x=[10−23]+s[3−121]+t[0112],s,t∈R\mathbf{x} = \begin{bmatrix} 1 \\ 0 \\ -2 \\ 3 \end{bmatrix} + s \begin{bmatrix} 3 \\ -1 \\ 2 \\ 1 \end{bmatrix} + t \begin{bmatrix} 0 \\ 1 \\ 1 \\ 2 \end{bmatrix}, \quad s, t \in \mathbb{R}

    • Parametric Equations:       {x1=1+3sx2=−s+tx3=−2+2s+tx4=3+s+2t\begin{cases} x_1 = 1 + 3s \\ x_2 = -s + t \\ x_3 = -2 + 2s + t \\ x_4 = 3 + s + 2t \end{cases}

  • Dimensional Comparison Table:

    • Line in Rn\mathbb{R}^n: Requires 11 non-zero direction vector; uses 11 parameter (tt).

    • Plane in Rn\mathbb{R}^n: Requires 22 non-zero, non-parallel direction vectors; uses 22 parameters (s,ts, t).

    • Note on ambient dimension: A plane in R4\mathbb{R}^4 requires four coordinates to specify position, but retains exactly two internal degrees of freedom (independent parameters). Cross product methods are strictly restricted to R3\mathbb{R}^3.

Classification of Lines in Three-Dimensional Space

  • The Four Geometric Relationships for Two Lines in R3\mathbb{R}^3:

    1. Intersecting: The lines possess non-parallel direction vectors and have exactly one common point of intersection.

    2. Parallel and Distinct: The direction vectors are scalar multiples of each other, but the lines share zero common points.

    3. Coincident (Same Line): The direction vectors are scalar multiples of each other, and all points are shared (every point on one line lies on the other).

    4. Skew: The direction vectors are not parallel (not scalar multiples), and the lines share zero common points. Skew lines lie in different parallel planes and never meet.

  • Geometric Visualizations of Spatial Lines:

    Skew lines lying in distinct parallel height levels
  • Key Geometric Warnings in Three Dimensions:

    • In R3\mathbb{R}^3, "not parallel" does not imply that two lines intersect (they may be skew).

    • In R3\mathbb{R}^3, "no intersection" does not imply that two lines are parallel (they may be skew).

    • Skew lines cannot exist in R2\mathbb{R}^2; two distinct lines in the plane are either parallel or intersecting.

    • A visual drawing is not an algebraic test for intersection: Two lines may appear to cross when projected onto a 2D drawing while actually passing each other at different depths or heights.

  • Systematic Classification Procedure:

    • Express the lines using distinct parameters:     ℓ1:x⃗=p⃗+su⃗,ℓ2:x⃗=q⃗+tv⃗\ell_1: \vec{x} = \vec{p} + s\vec{u}, \quad \ell_2: \vec{x} = \vec{q} + t\vec{v}

    • Step 1: Compare Direction Vectors (u⃗\vec{u} and v⃗\vec{v}):

    • Check whether u⃗=kv⃗\vec{u} = k\vec{v} for some non-zero scalar kk.

    • In R3\mathbb{R}^3, parallel directions can equivalently be checked by verifying u⃗×v⃗=0⃗\vec{u} \times \vec{v} = \vec{0}.

    • If directions are parallel:

      • Test whether the starting point QQ of line ℓ2\ell_2 lies on ℓ1\ell_1 by solving q⃗−p⃗=cu⃗\vec{q} - \vec{p} = c\vec{u}.

      • If a scalar cc exists: The lines are coincident.

      • If no scalar cc exists: The lines are parallel and distinct.

    • If directions are non-parallel:

      • Look for a common point by equating coordinates:         p⃗+su⃗=q⃗+tv⃗\vec{p} + s\vec{u} = \vec{q} + t\vec{v}

      • This produces a linear system of 3 equations in 2 variables (ss and tt):         {p1+su1=q1+tv1p2+su2=q2+tv2p3+su3=q3+tv3\begin{cases} p_1 + s u_1 = q_1 + t v_1 \\ p_2 + s u_2 = q_2 + t v_2 \\ p_3 + s u_3 = q_3 + t v_3 \end{cases}

      • Solve a convenient pair of equations for ss and tt, then substitute into the third remaining equation:

      • If the third equation is satisfied: The system has a unique solution pair (s,t)(s, t), and the lines are intersecting.

      • If the third equation produces a contradiction: The lines are skew.

    • Essential Caution: Never set s=ts = t. A shared intersection point may be reached at entirely different parameter values along each line.

  • Classification Case Studies:

    • Case 1: Intersecting Lines (Example 1):     ℓ1:x⃗=[102]+s[12−1],ℓ2:x⃗=[15−1]+t[2−11]\ell_1: \vec{x} = \begin{bmatrix} 1 \\ 0 \\ 2 \end{bmatrix} + s \begin{bmatrix} 1 \\ 2 \\ -1 \end{bmatrix}, \quad \ell_2: \vec{x} = \begin{bmatrix} 1 \\ 5 \\ -1 \end{bmatrix} + t \begin{bmatrix} 2 \\ -1 \\ 1 \end{bmatrix}

    • Directions: [12−1]T\begin{bmatrix} 1 & 2 & -1 \end{bmatrix}^T and [2−11]T\begin{bmatrix} 2 & -1 & 1 \end{bmatrix}^T are not scalar multiples (1/2≠2/(−1)1/2 \neq 2/(-1)).

    • Coordinate Equations:     {1+s=1+2t  ⟹  s=2t2s=5−t  ⟹  2(2t)=5−t  ⟹  5t=5  ⟹  t=12−s=−1+t\begin{cases} 1 + s = 1 + 2t & \implies s = 2t \\ 2s = 5 - t & \implies 2(2t) = 5 - t \implies 5t = 5 \implies t = 1 \\ 2 - s = -1 + t & \end{cases}

    • Since t=1t = 1, s=2(1)=2s = 2(1) = 2.

    • Check third equation: 2−(2)=02 - (2) = 0; −1+(1)=0-1 + (1) = 0. Both sides equal 00, satisfying the equation.

    • Find intersection point: Substitute s=2s = 2 into ℓ1\ell_1:     x⃗=[102]+2[12−1]=[340]\vec{x} = \begin{bmatrix} 1 \\ 0 \\ 2 \end{bmatrix} + 2 \begin{bmatrix} 1 \\ 2 \\ -1 \end{bmatrix} = \begin{bmatrix} 3 \\ 4 \\ 0 \end{bmatrix}

    • Verification: Substitute t=1t = 1 into ℓ2\ell_2: x⃗=[15−1]+1[2−11]=[340]\vec{x} = \begin{bmatrix} 1 \\ 5 \\ -1 \end{bmatrix} + 1 \begin{bmatrix} 2 \\ -1 \\ 1 \end{bmatrix} = \begin{bmatrix} 3 \\ 4 \\ 0 \end{bmatrix}.

    • Conclusion: Intersecting at point (3,4,0)(3, 4, 0).

  • Case 2: Parallel and Distinct Lines (Example 2):   ℓ1:x⃗=[102]+s[12−1],ℓ2:x⃗=[112]+t[24−2]\ell_1: \vec{x} = \begin{bmatrix} 1 \\ 0 \\ 2 \end{bmatrix} + s \begin{bmatrix} 1 \\ 2 \\ -1 \end{bmatrix}, \quad \ell_2: \vec{x} = \begin{bmatrix} 1 \\ 1 \\ 2 \end{bmatrix} + t \begin{bmatrix} 2 \\ 4 \\ -2 \end{bmatrix}

    • Directions: [24−2]T=2[12−1]T\begin{bmatrix} 2 & 4 & -2 \end{bmatrix}^T = 2 \begin{bmatrix} 1 & 2 & -1 \end{bmatrix}^T. Directions are parallel.

    • Check point Q=(1,1,2)Q = (1, 1, 2) on ℓ1\ell_1:     [112]=[102]+s[12−1]  ⟹  {1=1+s  ⟹  s=01=0+2s  ⟹  s=1/2\begin{bmatrix} 1 \\ 1 \\ 2 \end{bmatrix} = \begin{bmatrix} 1 \\ 0 \\ 2 \end{bmatrix} + s \begin{bmatrix} 1 \\ 2 \\ -1 \end{bmatrix} \implies \begin{cases} 1 = 1 + s \implies s = 0 \\ 1 = 0 + 2s \implies s = 1/2 \end{cases}

    • Inconsistent scalar ss (0≠1/20 \neq 1/2). Point QQ does not lie on ℓ1\ell_1.

    • Conclusion: Parallel and distinct.

  • Case 3: Coincident Lines (Example 3):   ℓ1:x⃗=[102]+s[12−1],ℓ2:x⃗=[340]+t[−2−42]\ell_1: \vec{x} = \begin{bmatrix} 1 \\ 0 \\ 2 \end{bmatrix} + s \begin{bmatrix} 1 \\ 2 \\ -1 \end{bmatrix}, \quad \ell_2: \vec{x} = \begin{bmatrix} 3 \\ 4 \\ 0 \end{bmatrix} + t \begin{bmatrix} -2 \\ -4 \\ 2 \end{bmatrix}

    • Directions: [−2−42]T=−2[12−1]T\begin{bmatrix} -2 & -4 & 2 \end{bmatrix}^T = -2 \begin{bmatrix} 1 & 2 & -1 \end{bmatrix}^T. Directions are parallel.

    • Check point Q=(3,4,0)Q = (3, 4, 0) on ℓ1\ell_1:     {3=1+s  ⟹  s=24=0+2s  ⟹  s=20=2−s  ⟹  s=2\begin{cases} 3 = 1 + s \implies s = 2 \\ 4 = 0 + 2s \implies s = 2 \\ 0 = 2 - s \implies s = 2 \end{cases}

    • Since s=2s = 2 satisfies all equations, QQ lies on ℓ1\ell_1.

    • Rewriting: q⃗+tv⃗=(p⃗+2u⃗)+t(−2u⃗)=p⃗+(2−2t)u⃗\vec{q} + t\vec{v} = (\vec{p} + 2\vec{u}) + t(-2\vec{u}) = \vec{p} + (2 - 2t)\vec{u}.

    • Conclusion: Coincident (infinitely many shared points).

  • Case 4: Skew Lines (Example 4):   ℓ1:x⃗=[102]+s[12−1],ℓ2:x⃗=[150]+t[2−11]\ell_1: \vec{x} = \begin{bmatrix} 1 \\ 0 \\ 2 \end{bmatrix} + s \begin{bmatrix} 1 \\ 2 \\ -1 \end{bmatrix}, \quad \ell_2: \vec{x} = \begin{bmatrix} 1 \\ 5 \\ 0 \end{bmatrix} + t \begin{bmatrix} 2 \\ -1 \\ 1 \end{bmatrix}

    • Directions: Non-parallel.

    • Coordinate Equations:     {1+s=1+2t  ⟹  s=2t2s=5−t  ⟹  2(2t)=5−t  ⟹  t=1,s=22−s=0+t\begin{cases} 1 + s = 1 + 2t & \implies s = 2t \\ 2s = 5 - t & \implies 2(2t) = 5 - t \implies t = 1, s = 2 \\ 2 - s = 0 + t & \end{cases}

    • Check third equation: Left-hand side is 2−s=2−2=02 - s = 2 - 2 = 0. Right-hand side is 0+t=10 + t = 1. Since 0≠10 \neq 1, the system has no solution.

    • Matching xx and yy coordinates is insufficient; all three coordinates must match simultaneously.

    • Conclusion: Skew.

  • Quick Classification Reference Checks (Against line ℓ:x⃗=(s,0,0)T\ell: \vec{x} = (s, 0, 0)^T):

    • (a) x⃗=(0,t,0)T\vec{x} = (0, t, 0)^T: Directions (1,0,0)T(1, 0, 0)^T and (0,1,0)T(0, 1, 0)^T are non-parallel. Equating coordinates yields s=0,t=0s = 0, t = 0. Intersecting at point (0,0,0)(0, 0, 0).

    • (b) x⃗=(0,t,1)T\vec{x} = (0, t, 1)^T: Directions are non-parallel. Equating zz-coordinates gives 0=10 = 1 (impossible). Skew.

    • (c) x⃗=(0,1,0)T+t(2,0,0)T\vec{x} = (0, 1, 0)^T + t(2, 0, 0)^T: Direction (2,0,0)T=2(1,0,0)T(2, 0, 0)^T = 2(1, 0, 0)^T (parallel). Point (0,1,0)(0, 1, 0) has y=1≠0y = 1 \neq 0, so it does not lie on ℓ\ell. Parallel and distinct.

  • (d) x⃗=(2,0,0)T+t(−1,0,0)T\vec{x} = (2, 0, 0)^T + t(-1, 0, 0)^T: Direction (−1,0,0)T=−1(1,0,0)T(-1, 0, 0)^T = -1(1, 0, 0)^T (parallel). Point (2,0,0)(2, 0, 0) lies on ℓ\ell at s=2s = 2. Coincident.

Orthogonal Projections and Distance Calculations

  • Vector Orthogonal Projection Formula:

    • For a non-zero vector w≠0\mathbf{w} \neq \mathbf{0}, the orthogonal projection of vector v\mathbf{v} onto w\mathbf{w} is:     projw(v)=v⋅ww⋅ww\text{proj}_\mathbf{w}(\mathbf{v}) = \frac{\mathbf{v} \cdot \mathbf{w}}{\mathbf{w} \cdot \mathbf{w}} \mathbf{w}

    • Decomposition:

    • Parallel component: projw(v)\text{proj}_\mathbf{w}(\mathbf{v}) is directed parallel to w\mathbf{w}.

    • Perpendicular component: The remainder v−projw(v)\mathbf{v} - \text{proj}_\mathbf{w}(\mathbf{v}) is strictly perpendicular to w\mathbf{w}.

    • Orthogonality check: [v−projw(v)]⋅w=v⋅w−v⋅ww⋅w(w⋅w)=0[\mathbf{v} - \text{proj}_\mathbf{w}(\mathbf{v})] \cdot \mathbf{w} = \mathbf{v} \cdot \mathbf{w} - \frac{\mathbf{v} \cdot \mathbf{w}}{\mathbf{w} \cdot \mathbf{w}}(\mathbf{w} \cdot \mathbf{w}) = 0.

  • Distance and Closest Point from a Point to a Line:

    Right triangle vector orthogonal projection decomposition
    • Let line ℓ\ell pass through base point AA with direction d≠0\mathbf{d} \neq \mathbf{0}, and let BB be an external point.

    • Total displacement vector: AB⃗=B−A\vec{AB} = B - A.

    • Displacement decomposition:     AB⃗=AH⃗+HB⃗\vec{AB} = \vec{AH} + \vec{HB}

    • Parallel displacement along the line: AH⃗=projd(AB⃗)\vec{AH} = \text{proj}_\mathbf{d}(\vec{AB})

    • Perpendicular displacement to the line: HB⃗=AB⃗−projd(AB⃗)\vec{HB} = \vec{AB} - \text{proj}_\mathbf{d}(\vec{AB})

    • Closest Point (HH): Obtained by starting at point AA and traversing the parallel displacement:     H=A+projd(AB⃗)H = A + \text{proj}_\mathbf{d}(\vec{AB})

    • Shortest Distance: The Euclidean norm of the perpendicular displacement vector:     dist(B,ℓ)=∥AB⃗−projd(AB⃗)∥\text{dist}(B, \ell) = \lVert \vec{AB} - \text{proj}_\mathbf{d}(\vec{AB}) \rVert

    • Category Distinction: The projection projd(AB⃗)\text{proj}_\mathbf{d}(\vec{AB}) is a vector; the closest location HH is a spatial point; the distance dist(B,ℓ)\text{dist}(B, \ell) is a non-negative scalar.

  • Worked Line Projection Problem (Example 19):

    • Find the distance from point B=(1,0,2)B = (1, 0, 2) to the line ℓ:x=[311]+t[−110]\ell: \mathbf{x} = \begin{bmatrix} 3 \\ 1 \\ 1 \end{bmatrix} + t \begin{bmatrix} -1 \\ 1 \\ 0 \end{bmatrix}.

    • Step 1: Displacement vector AB⃗=B−A\vec{AB} = B - A:     AB⃗=[102]−[311]=[−2−11]\vec{AB} = \begin{bmatrix} 1 \\ 0 \\ 2 \end{bmatrix} - \begin{bmatrix} 3 \\ 1 \\ 1 \end{bmatrix} = \begin{bmatrix} -2 \\ -1 \\ 1 \end{bmatrix}

    • Step 2: Projection of AB⃗\vec{AB} onto d=[−110]T\mathbf{d} = \begin{bmatrix} -1 & 1 & 0 \end{bmatrix}^T:     AB⃗⋅d=(−2)(−1)+(−1)(1)+(1)(0)=2−1+0=1\vec{AB} \cdot \mathbf{d} = (-2)(-1) + (-1)(1) + (1)(0) = 2 - 1 + 0 = 1     d⋅d=(−1)2+12+02=2\mathbf{d} \cdot \mathbf{d} = (-1)^2 + 1^2 + 0^2 = 2     projd(AB⃗)=12[−110]=[−1/21/20]\text{proj}_\mathbf{d}(\vec{AB}) = \frac{1}{2} \begin{bmatrix} -1 \\ 1 \\ 0 \end{bmatrix} = \begin{bmatrix} -1/2 \\ 1/2 \\ 0 \end{bmatrix}

    • Step 3: Closest point HH:     H=A+projd(AB⃗)=[311]+[−1/21/20]=(52,32,1)H = A + \text{proj}_\mathbf{d}(\vec{AB}) = \begin{bmatrix} 3 \\ 1 \\ 1 \end{bmatrix} + \begin{bmatrix} -1/2 \\ 1/2 \\ 0 \end{bmatrix} = \left( \frac{5}{2}, \frac{3}{2}, 1 \right)

    • Step 4: Perpendicular remainder and distance:     HB⃗=AB⃗−projd(AB⃗)=[−2−11]−[−1/21/20]=[−3/2−3/21]\vec{HB} = \vec{AB} - \text{proj}_\mathbf{d}(\vec{AB}) = \begin{bmatrix} -2 \\ -1 \\ 1 \end{bmatrix} - \begin{bmatrix} -1/2 \\ 1/2 \\ 0 \end{bmatrix} = \begin{bmatrix} -3/2 \\ -3/2 \\ 1 \end{bmatrix}     dist(B,ℓ)=∥HB⃗∥=(−32)2+(−32)2+12=94+94+44=224=222\text{dist}(B, \ell) = \lVert \vec{HB} \rVert = \sqrt{\left(-\frac{3}{2}\right)^2 + \left(-\frac{3}{2}\right)^2 + 1^2} = \sqrt{\frac{9}{4} + \frac{9}{4} + \frac{4}{4}} = \sqrt{\frac{22}{4}} = \frac{\sqrt{22}}{2}

  • Distance and Closest Point from a Point to a Plane:

    Vector orthogonal projection onto a plane normal
    • Let AA be any chosen point lying on plane P\mathcal{P}, let n≠0\mathbf{n} \neq \mathbf{0} be a vector normal (perpendicular) to P\mathcal{P}, and let BB be an external point.

    • Projection onto Normal: Projecting the displacement vector AB⃗\vec{AB} directly onto n\mathbf{n} yields the perpendicular displacement vector connecting HH to BB:     HB⃗=projn(AB⃗)\vec{HB} = \text{proj}_\mathbf{n}(\vec{AB})

    • Shortest Distance:     dist(B,P)=∥HB⃗∥=∥projn(AB⃗)∥\text{dist}(B, \mathcal{P}) = \lVert \vec{HB} \rVert = \lVert \text{proj}_\mathbf{n}(\vec{AB}) \rVert

    • Closest Point (HH):     H=B−projn(AB⃗)H = B - \text{proj}_\mathbf{n}(\vec{AB})

    • Procedural Contrast: For a line, project onto the direction vector and subtract to get the perpendicular component. For a plane, projecting onto the normal vector immediately yields the perpendicular component directly.

  • Worked Plane Projection Problem (Example 20):

    • Find the distance from point B=(1,0,2)B = (1, 0, 2) to plane P:x+y−z=1\mathcal{P}: x + y - z = 1, and determine the closest point HH.

    • Identify normal vector from coefficients: n=[11−1]T\mathbf{n} = \begin{bmatrix} 1 & 1 & -1 \end{bmatrix}^T.

    • Select convenient point A∈PA \in \mathcal{P}: Choose y=0,z=0  ⟹  x=1y = 0, z = 0 \implies x = 1, giving A=(1,0,0)A = (1, 0, 0).

    • Displacement vector AB⃗=B−A\vec{AB} = B - A:     AB⃗=[102]−[100]=[002]\vec{AB} = \begin{bmatrix} 1 \\ 0 \\ 2 \end{bmatrix} - \begin{bmatrix} 1 \\ 0 \\ 0 \end{bmatrix} = \begin{bmatrix} 0 \\ 0 \\ 2 \end{bmatrix}

    • Projection onto normal n\mathbf{n}:     AB⃗⋅n=0(1)+0(1)+2(−1)=−2\vec{AB} \cdot \mathbf{n} = 0(1) + 0(1) + 2(-1) = -2     n⋅n=12+12+(−1)2=3\mathbf{n} \cdot \mathbf{n} = 1^2 + 1^2 + (-1)^2 = 3     projn(AB⃗)=−23[11−1]=[−2/3−2/32/3]\text{proj}_\mathbf{n}(\vec{AB}) = \frac{-2}{3} \begin{bmatrix} 1 \\ 1 \\ -1 \end{bmatrix} = \begin{bmatrix} -2/3 \\ -2/3 \\ 2/3 \end{bmatrix}

    • Shortest Distance:     dist(B,P)=∥projn(AB⃗)∥=(−23)2+(−23)2+(23)2=4+4+49=123=233\text{dist}(B, \mathcal{P}) = \lVert \text{proj}_\mathbf{n}(\vec{AB}) \rVert = \sqrt{\left(-\frac{2}{3}\right)^2 + \left(-\frac{2}{3}\right)^2 + \left(\frac{2}{3}\right)^2} = \sqrt{\frac{4 + 4 + 4}{9}} = \frac{\sqrt{12}}{3} = \frac{2\sqrt{3}}{3}

    • Closest Point HH:     H=B−projn(AB⃗)=[102]−[−2/3−2/32/3]=(53,23,43)H = B - \text{proj}_\mathbf{n}(\vec{AB}) = \begin{bmatrix} 1 \\ 0 \\ 2 \end{bmatrix} - \begin{bmatrix} -2/3 \\ -2/3 \\ 2/3 \end{bmatrix} = \left( \frac{5}{3}, \frac{2}{3}, \frac{4}{3} \right)

    • Verification: Substitute HH into the plane equation:     x+y−z=53+23−43=33=1x + y - z = \frac{5}{3} + \frac{2}{3} - \frac{4}{3} = \frac{3}{3} = 1     The point strictly satisfies the plane equation.

Vector Spans and Spanning Sets

  • Linear Combinations of Vectors:

    • Standard basis vectors in R2\mathbb{R}^2:     e1=[10],e2=[01]\mathbf{e}_1 = \begin{bmatrix} 1 \\ 0 \end{bmatrix}, \quad \mathbf{e}_2 = \begin{bmatrix} 0 \\ 1 \end{bmatrix}

    • Any vector in R2\mathbb{R}^2 decomposes as a unique combination:     [xy]=xe1+ye2\begin{bmatrix} x \\ y \end{bmatrix} = x \mathbf{e}_1 + y \mathbf{e}_2

    • Key Connection: Asking whether vector b\mathbf{b} is a linear combination of vectors v1,v2,…,vk\mathbf{v}_1, \mathbf{v}_2, \dots, \mathbf{v}_k is equivalent to determining whether the vector equation:     c1v1+c2v2+⋯+ckvk=bc_1 \mathbf{v}_1 + c_2 \mathbf{v}_2 + \dots + c_k \mathbf{v}_k = \mathbf{b}     possesses a consistent solution for the unknown scalars c1,c2,…,ckc_1, c_2, \dots, c_k.

  • Consistency Test for Linear Combinations:

    • Construct matrix AA using the given vectors as columns:     A=[v1v2…vk]A = \begin{bmatrix} \mathbf{v}_1 & \mathbf{v}_2 & \dots & \mathbf{v}_k \end{bmatrix}

    • The vector equation is equivalent to the matrix-vector system Ac=bA\mathbf{c} = \mathbf{b}.

    • Theorem: The system Ac=bA\mathbf{c} = \mathbf{b} is consistent if and only if b\mathbf{b} can be written as a linear combination of the columns of AA.

  • Worked Linear Combination Problem (Example 7):

    • Determine whether b=[2−2−1]\mathbf{b} = \begin{bmatrix} 2 \\ -2 \\ -1 \end{bmatrix} is a linear combination of:     v1=[10−1],v2=[2−31],v3=[5−40]\mathbf{v}_1 = \begin{bmatrix} 1 \\ 0 \\ -1 \end{bmatrix}, \quad \mathbf{v}_2 = \begin{bmatrix} 2 \\ -3 \\ 1 \end{bmatrix}, \quad \mathbf{v}_3 = \begin{bmatrix} 5 \\ -4 \\ 0 \end{bmatrix}

    • Set up augmented matrix and row reduce:     [12520−3−4−2−110−1]\left[\begin{array}{ccc|c} 1 & 2 & 5 & 2 \\ 0 & -3 & -4 & -2 \\ -1 & 1 & 0 & -1 \end{array}\right]

    • Perform R3←R3+R1R_3 \leftarrow R_3 + R_1:     [12520−3−4−20351]\left[\begin{array}{ccc|c} 1 & 2 & 5 & 2 \\ 0 & -3 & -4 & -2 \\ 0 & 3 & 5 & 1 \end{array}\right]

    • Perform R3←R3+R2R_3 \leftarrow R_3 + R_2:     [12520−3−4−2001−1]\left[\begin{array}{ccc|c} 1 & 2 & 5 & 2 \\ 0 & -3 & -4 & -2 \\ 0 & 0 & 1 & -1 \end{array}\right]

    • Back substitution:

    • Row 3: c3=−1c_3 = -1.

    • Row 2: −3c2−4(−1)=−2  ⟹  −3c2=−6  ⟹  c2=2-3c_2 - 4(-1) = -2 \implies -3c_2 = -6 \implies c_2 = 2.

    • Row 1: c1+2(2)+5(−1)=2  ⟹  c1−1=2  ⟹  c1=3c_1 + 2(2) + 5(-1) = 2 \implies c_1 - 1 = 2 \implies c_1 = 3.

    • The system is consistent. Solution: b=3v1+2v2−v3\mathbf{b} = 3\mathbf{v}_1 + 2\mathbf{v}_2 - \mathbf{v}_3.

  • Testing Multiple Vectors (Example 8):

    • Let u=[103],v=[−11−3]\mathbf{u} = \begin{bmatrix} 1 \\ 0 \\ 3 \end{bmatrix}, \mathbf{v} = \begin{bmatrix} -1 \\ 1 \\ -3 \end{bmatrix}.

    • (a) Testing b1=[123]T\mathbf{b}_1 = \begin{bmatrix} 1 & 2 & 3 \end{bmatrix}^T:     [1−110123−33]→R3←R3−3R1[1−11012000]\left[\begin{array}{cc|c} 1 & -1 & 1 \\ 0 & 1 & 2 \\ 3 & -3 & 3 \end{array}\right] \xrightarrow{R_3 \leftarrow R_3 - 3R_1} \left[\begin{array}{cc|c} 1 & -1 & 1 \\ 0 & 1 & 2 \\ 0 & 0 & 0 \end{array}\right]     Consistent. c2=2c_2 = 2, c1−2=1  ⟹  c1=3c_1 - 2 = 1 \implies c_1 = 3. Thus, b1=3u+2v\mathbf{b}_1 = 3\mathbf{u} + 2\mathbf{v}.

    • (b) Testing b2=[234]T\mathbf{b}_2 = \begin{bmatrix} 2 & 3 & 4 \end{bmatrix}^T:     [1−120133−34]→R3←R3−3R1[1−1201300−2]\left[\begin{array}{cc|c} 1 & -1 & 2 \\ 0 & 1 & 3 \\ 3 & -3 & 4 \end{array}\right] \xrightarrow{R_3 \leftarrow R_3 - 3R_1} \left[\begin{array}{cc|c} 1 & -1 & 2 \\ 0 & 1 & 3 \\ 0 & 0 & -2 \end{array}\right]     Inconsistent (0=−20 = -2). Vector b2\mathbf{b}_2 cannot be written as a linear combination of u\mathbf{u} and v\mathbf{v}.

  • Definition of Span (Definition 8):

    • Let S={v1,v2,…,vk}⊆RnS = \{\mathbf{v}_1, \mathbf{v}_2, \dots, \mathbf{v}_k\} \subseteq \mathbb{R}^n.

    • The set of all linear combinations of v1,…,vk\mathbf{v}_1, \dots, \mathbf{v}_k is called the span of SS:     span(S)={c1v1+⋯+ckvk:c1,…,ck∈R}\text{span}(S) = \{c_1 \mathbf{v}_1 + \dots + c_k \mathbf{v}_k : c_1, \dots, c_k \in \mathbb{R}\}

    • If span(S)=Rn\text{span}(S) = \mathbb{R}^n, then SS is designated a spanning set for Rn\mathbb{R}^n.

    • Checkpoint: Span answers the core question: Which vectors can be reached using linear combinations of the given vectors?

  • Proving a Set Spans R2\mathbb{R}^2 (Example 9):

    • Show that span{[2−1],[13]}=R2\text{span}\left\{ \begin{bmatrix} 2 \\ -1 \end{bmatrix}, \begin{bmatrix} 1 \\ 3 \end{bmatrix} \right\} = \mathbb{R}^2.

    • Let [ab]∈R2\begin{bmatrix} a \\ b \end{bmatrix} \in \mathbb{R}^2 be arbitrary. Solve c1[2−1]+c2[13]=[ab]c_1 \begin{bmatrix} 2 \\ -1 \end{bmatrix} + c_2 \begin{bmatrix} 1 \\ 3 \end{bmatrix} = \begin{bmatrix} a \\ b \end{bmatrix}:     [21a−13b]→R1↔R2[−13b21a]→R2←R2+2R1[−13b07a+2b]\left[\begin{array}{cc|c} 2 & 1 & a \\ -1 & 3 & b \end{array}\right] \xrightarrow{R_1 \leftrightarrow R_2} \left[\begin{array}{cc|c} -1 & 3 & b \\ 2 & 1 & a \end{array}\right] \xrightarrow{R_2 \leftarrow R_2 + 2R_1} \left[\begin{array}{cc|c} -1 & 3 & b \\ 0 & 7 & a + 2b \end{array}\right]

    • Because there is a pivot in every row of the coefficient matrix (7≠07 \neq 0), the system is consistent for any choices of real numbers aa and bb.

    • Specifically, c2=a+2b7c_2 = \frac{a + 2b}{7} and c1=3c2−b=3a−b7c_1 = 3c_2 - b = \frac{3a - b}{7}.

    • Therefore, the two vectors span the entirety of R2\mathbb{R}^2.

  • Geometric Description of Span (Example 10):

    • Describe span{[103],[−11−3]}\text{span}\left\{ \begin{bmatrix} 1 \\ 0 \\ 3 \end{bmatrix}, \begin{bmatrix} -1 \\ 1 \\ -3 \end{bmatrix} \right\}.

    • The two vectors are not scalar multiples of each other, meaning they define two distinct directions.

    • The span is not all of R3\mathbb{R}^3 because two vectors can span at most two dimensions.

    • Geometrically, the span is a two-dimensional flat plane passing through the origin in R3\mathbb{R}^3.

  • Inclusion of the Zero Vector:

    • The zero vector 0\mathbf{0} is always present in the span of any non-empty set of vectors.

    • Reason: Simply assign every scalar weight to zero: 0v1+0v2+⋯+0vk=00\mathbf{v}_1 + 0\mathbf{v}_2 + \dots + 0\mathbf{v}_k = \mathbf{0}.

    • Connection: This parallels the property that homogeneous linear systems Ax=0A\mathbf{x} = \mathbf{0} are always consistent because the trivial solution x=0\mathbf{x} = \mathbf{0} is always valid.

Linear Independence and Rank Criteria

  • Definition of Linear Independence (Definition 9):

    • A set of vectors S={v1,…,vk}⊆RnS = \{\mathbf{v}_1, \dots, \mathbf{v}_k\} \subseteq \mathbb{R}^n is linearly independent if the vector equation:     c1v1+c2v2+⋯+ckvk=0c_1 \mathbf{v}_1 + c_2 \mathbf{v}_2 + \dots + c_k \mathbf{v}_k = \mathbf{0}     forces all scalar coefficients to be zero:     c1=c2=⋯=ck=0c_1 = c_2 = \dots = c_k = 0

    • If there exists a solution where at least one coefficient is non-zero, the set is linearly dependent.

    • In a linearly dependent set, at least one vector can be expressed as a linear combination of the other vectors.

    • Zero Vector Rule: Any set containing the zero vector 0\mathbf{0} is automatically linearly dependent.

    • Justification: If v1=0\mathbf{v}_1 = \mathbf{0}, then 1v1+0v2+⋯+0vk=01\mathbf{v}_1 + 0\mathbf{v}_2 + \dots + 0\mathbf{v}_k = \mathbf{0} provides a non-trivial solution (coefficient c1=1≠0c_1 = 1 \neq 0).

  • Geometric and Structural Significance:

    • If a vector is already within the span of a set of vectors (as in Example 8(a)), appending it to the set adds no new directional coverage; the span remains unchanged and the set becomes dependent.

    • If a vector is not in the span (as in Example 8(b)), appending it adds an entirely new dimension/direction to the span, and linear independence is preserved.

  • Testing Methods for Linear Independence:

    1. Homogeneous-System Test:

    • Arrange vectors as the columns of matrix A=[v1v2…vk]A = \begin{bmatrix} \mathbf{v}_1 & \mathbf{v}_2 & \dots & \mathbf{v}_k \end{bmatrix}.

    • The vectors are linearly independent if and only if the homogeneous system Ac=0A\mathbf{c} = \mathbf{0} has exclusively the trivial solution c=0\mathbf{c} = \mathbf{0}.

    • If free variables exist (non-trivial solutions exist), the vectors are linearly dependent.

    1. Row-Vector Rank Test:

    • Arrange kk vectors in Rn\mathbb{R}^n as the rows of a k×nk \times n matrix:        A=[v1Tv2T⋮vkT]A = \begin{bmatrix} \mathbf{v}_1^T \\ \mathbf{v}_2^T \\ \vdots \\ \mathbf{v}_k^T \end{bmatrix}

    • The row vectors are linearly independent if and only if rank(A)=k\text{rank}(A) = k (meaning no all-zero rows appear after reduction to REF).

    • By tracking row operations, a specific dependence relation among dependent rows can be identified directly from the combination producing an all-zero row.

    1. Quick Dimension Test:

    • Any set of kk vectors in Rn\mathbb{R}^n is guaranteed to be linearly dependent if k>nk > n.

    • Justification: Placing the vectors as columns of an n×kn \times k matrix results in fewer rows than columns (n<kn < k). The rank cannot exceed nn. Thus, the number of pivots is at most n<kn < k, which guarantees at least k−nk - n free variables in the homogeneous system Ac=0A\mathbf{c} = \mathbf{0}.

  • Worked Linear Independence Examples:

    • Example 11(a): S={[14],[−12]}S = \left\{ \begin{bmatrix} 1 \\ 4 \end{bmatrix}, \begin{bmatrix} -1 \\ 2 \end{bmatrix} \right\}

    • Solve c1[14]+c2[−12]=[00]c_1 \begin{bmatrix} 1 \\ 4 \end{bmatrix} + c_2 \begin{bmatrix} -1 \\ 2 \end{bmatrix} = \begin{bmatrix} 0 \\ 0 \end{bmatrix}.

    • Augmented matrix:       [1−10420]→R2←R2−4R1[1−10060]\left[\begin{array}{cc|c} 1 & -1 & 0 \\ 4 & 2 & 0 \end{array}\right] \xrightarrow{R_2 \leftarrow R_2 - 4R_1} \left[\begin{array}{cc|c} 1 & -1 & 0 \\ 0 & 6 & 0 \end{array}\right]

    • Two pivots, zero free variables. Only c1=c2=0c_1 = c_2 = 0 works. The set is linearly independent.

    • Example 11(b): S={[110],[011],[101]}S = \left\{ \begin{bmatrix} 1 \\ 1 \\ 0 \end{bmatrix}, \begin{bmatrix} 0 \\ 1 \\ 1 \end{bmatrix}, \begin{bmatrix} 1 \\ 0 \\ 1 \end{bmatrix} \right\}

    • Augmented matrix:       [101011000110]→R2←R2−R1[101001−100110]→R3←R3−R2[101001−100020]\left[\begin{array}{ccc|c} 1 & 0 & 1 & 0 \\ 1 & 1 & 0 & 0 \\ 0 & 1 & 1 & 0 \end{array}\right] \xrightarrow{R_2 \leftarrow R_2 - R_1} \left[\begin{array}{ccc|c} 1 & 0 & 1 & 0 \\ 0 & 1 & -1 & 0 \\ 0 & 1 & 1 & 0 \end{array}\right] \xrightarrow{R_3 \leftarrow R_3 - R_2} \left[\begin{array}{ccc|c} 1 & 0 & 1 & 0 \\ 0 & 1 & -1 & 0 \\ 0 & 0 & 2 & 0 \end{array}\right]

    • Three pivots, rank equals 33. No free variables. Only the trivial solution exists. The set is linearly independent.

    • Example 11(c) & Example 13 (Row-Vector Rank Test):

    • Testing vectors as rows in matrix AA:       A=[12011−1142]A = \begin{bmatrix} 1 & 2 & 0 \\ 1 & 1 & -1 \\ 1 & 4 & 2 \end{bmatrix}

    • Row reduce: R2←R2−R1R_2 \leftarrow R_2 - R_1 and R3←R3−R1R_3 \leftarrow R_3 - R_1:       [1200−1−1022]\begin{bmatrix} 1 & 2 & 0 \\ 0 & -1 & -1 \\ 0 & 2 & 2 \end{bmatrix}

    • Perform R3←R3+2R2R_3 \leftarrow R_3 + 2R_2:       [1200−1−1000]\begin{bmatrix} 1 & 2 & 0 \\ 0 & -1 & -1 \\ 0 & 0 & 0 \end{bmatrix}

    • Here, rank(A)=2<3\text{rank}(A) = 2 < 3 (a row of zeros appears). The rows are linearly dependent.

    • Relation tracking: Notice R3−R1+2(R2−R1)=0  ⟹  R3+2R2−3R1=0R_3 - R_1 + 2(R_2 - R_1) = 0 \implies R_3 + 2R_2 - 3R_1 = 0.

    • Linear relation: v3=3v1−2v2\mathbf{v}_3 = 3\mathbf{v}_1 - 2\mathbf{v}_2.

    • Example 12 (Standard Basis Independence):

    • Testing standard basis vectors e1,e2,…,en\mathbf{e}_1, \mathbf{e}_2, \dots, \mathbf{e}_n using columns of AA:       A=[e1e2…en]=InA = \begin{bmatrix} \mathbf{e}_1 & \mathbf{e}_2 & \dots & \mathbf{e}_n \end{bmatrix} = I_n

    • The homogeneous system is Inc=0I_n \mathbf{c} = \mathbf{0}.

    • Since Inc=cI_n \mathbf{c} = \mathbf{c}, this immediately forces c=0\mathbf{c} = \mathbf{0}.

    • The standard basis vectors are always linearly independent.

    • Example 14 (Applying the Quick Dimension Test):

    • Set: S={[13],[24],[31]}S = \left\{ \begin{bmatrix} 1 \\ 3 \end{bmatrix}, \begin{bmatrix} 2 \\ 4 \end{bmatrix}, \begin{bmatrix} 3 \\ 1 \end{bmatrix} \right\}.

    • The set contains k=3k = 3 vectors residing in Rn\mathbb{R}^n with n=2n = 2.

    • Since k=3>2=nk = 3 > 2 = n, by the Quick Dimension Test, the set is guaranteed to be linearly dependent without needing computation.

    • Row reduction verification:       [123341]→R2←R2−3R1[1230−2−8]\begin{bmatrix} 1 & 2 & 3 \\ 3 & 4 & 1 \end{bmatrix} \xrightarrow{R_2 \leftarrow R_2 - 3R_1} \begin{bmatrix} 1 & 2 & 3 \\ 0 & -2 & -8 \end{bmatrix}       Two pivots with 3 columns leaves 3−2=13 - 2 = 1 free variable, confirming non-trivial solutions and linear dependence.

Questions & Discussion

  • Prompt: Can one vector by itself generate every vector in R2\mathbb{R}^2 by scalar multiplication?

    • Response: No. The scalar multiples of a single non-zero vector v∈R2\mathbf{v} \in \mathbb{R}^2 form the set {cv:c∈R}\{c\mathbf{v} : c \in \mathbb{R}\}, which geometrically traces only a one-dimensional line through the origin. It cannot cover the entire two-dimensional area of R2\mathbb{R}^2.

  • Prompt: Do any two vectors in R2\mathbb{R}^2 always generate every vector in R2\mathbb{R}^2?

    • Response: No. If the two vectors are collinear (scalar multiples of one another) or if one/both are the zero vector, their linear combinations only span a line or the origin, rather than all of R2\mathbb{R}^2. To generate all of R2\mathbb{R}^2, the two vectors must be non-parallel (linearly independent).

  • Prompt: Which familiar vectors are guaranteed to span Rn\mathbb{R}^n?

    • Response: The standard basis vectors e1,e2,…,en\mathbf{e}_1, \mathbf{e}_2, \dots, \mathbf{e}_n (where ej\mathbf{e}_j has a 11 in the jj-th entry and 00 elsewhere) are guaranteed to span Rn\mathbb{R}^n, since any arbitrary vector x=[x1x2…xn]T\mathbf{x} = \begin{bmatrix} x_1 & x_2 & \dots & x_n \end{bmatrix}^T can be expressed as x=x1e1+x2e2+⋯+xnen\mathbf{x} = x_1\mathbf{e}_1 + x_2\mathbf{e}_2 + \dots + x_n\mathbf{e}_n.

  • Prompt: Why is 0\mathbf{0} always in the span of any non-empty set of vectors? What idea from homogeneous systems does this recall?

    • Response: The zero vector is always present because setting all scalar weights to zero in a linear combination yields 0v1+0v2+⋯+0vk=00\mathbf{v}_1 + 0\mathbf{v}_2 + \dots + 0\mathbf{v}_k = \mathbf{0}. This mirrors the property of homogeneous systems Ax=0A\mathbf{x} = \mathbf{0}, which are always consistent because they unconditionally admit the trivial solution x=0\mathbf{x} = \mathbf{0}.

  • Prompt: What does the result of Example 8(a) tell us about the three vectors together? Does the third vector add a new direction to the span?

    • Response: Since [123]T\begin{bmatrix} 1 & 2 & 3 \end{bmatrix}^T is already a linear combination of u\mathbf{u} and v\mathbf{v}, it lies completely within their flat two-dimensional spanning plane. Adding this third vector to the set introduces redundancy, provides no new geometric direction, and renders the three vectors linearly dependent.

  • Prompt: What happens when the vector from Example 8(b) is added to the set?

    • Response: Because [234]T\begin{bmatrix} 2 & 3 & 4 \end{bmatrix}^T cannot be written as a linear combination of u\mathbf{u} and v\mathbf{v}, it points outside their two-dimensional plane. Appending it adds a third independent direction, expanding the span from a 2D plane to all of R3\mathbb{R}^3, and keeping the three vectors linearly independent.

  • Prompt: We have used augmented matrices to organize linear systems. What is a matrix, and where have we seen one before?

    • Response: A matrix is a rectangular mathematical array of numbers organized into rows and columns. Beyond augmented matrices, matrices appear as coefficient matrices of linear systems, coordinate transformations, and data grids.

  • Prompt: What should we check before multiplying partitioned matrices?

    • Response: We must verify conformity between blocks: the column partition sizes of the first matrix must match the row partition sizes of the second matrix so that every required submatrix product and addition is dimensionally defined.

  • Prompt: For real numbers, ab=0ab = 0 implies a=0a = 0 or b=0b = 0. Does AB=OAB = O force A=OA = O or B=OB = O for matrices? Must ABAB equal BABA?

    • Response: No to both questions. Two non-zero matrices can multiply to yield the zero matrix OO, meaning matrix arithmetic contains non-zero divisors of zero. Furthermore, matrix multiplication is non-commutative: ABAB does not generally equal BABA, and one product may exist even if the other is undefined.

  • Prompt: If we have AX=BAX = B, can we simply "divide by AA"? What extra concept is needed to solve such an equation?

    • Response: Division by a matrix is undefined. To isolate XX, we need the concept of a multiplicative matrix inverse, denoted A−1A^{-1}. If AA is square and invertible, we multiply both sides on the left by A−1A^{-1}: A−1AX=A−1B  ⟹  X=A−1BA^{-1}AX = A^{-1}B \implies X = A^{-1}B. If AA is not square or has no inverse, this technique cannot be applied.

  • Prompt: What does the word linear mean? What features should a linear equation have?

    • Response: In algebra, linear signifies that every variable appears only to the first power, is not multiplied or divided by another variable, and does not appear inside transcendental, exponential, or root functions. Geometrically, the solution sets of linear equations form flat geometric objects (lines, planes, or hyperplanes) without curves, kinks, or self-intersections.

  • Prompt: Why can a linear system not have exactly two solutions?

    • Response: Geometrically, solutions represent intersections of flat planes or lines. If two flat geometric surfaces intersect at two distinct points, they must intersect along the entire straight line connecting those two points, yielding infinitely many solutions. Algebraically, if u\mathbf{u} and v\mathbf{v} are distinct solutions to Ax=bA\mathbf{x} = \mathbf{b}, then any convex combination xt=(1−t)u+tv\mathbf{x}_t = (1 - t)\mathbf{u} + t\mathbf{v} for t∈Rt \in \mathbb{R} satisfies:     Axt=(1−t)Au+tAv=(1−t)b+tb=bA\mathbf{x}_t = (1 - t)A\mathbf{u} + tA\mathbf{v} = (1 - t)\mathbf{b} + t\mathbf{b} = \mathbf{b}     This generates infinitely many distinct solutions, rendering exactly two solutions mathematically impossible.

  • Prompt: What solution is every homogeneous system guaranteed to have? Can a homogeneous system have no solutions?

    • Response: Every homogeneous system is guaranteed to have the trivial zero solution x=0\mathbf{x} = \mathbf{0}. Because the zero vector is always a valid solution, a homogeneous system can never be inconsistent (it can never have zero solutions).

  • Prompt: If a homogeneous system has mm equations and nn variables with m<nm < n, what can be said about the number of solutions?

    • Response: It is guaranteed to have infinitely many solutions. Because the rank is bounded by the number of equations (rank(A)≤m<n\text{rank}(A) \le m < n), the system must have at least n−m>0n - m > 0 free variables, producing infinitely many non-trivial solutions alongside the trivial solution.