Comprehensive Exercises

I. Determinants

Problem 1

Compute the determinant D=∣1234234134124123∣D = \begin{vmatrix} 1 & 2 & 3 & 4 \\ 2 & 3 & 4 & 1 \\ 3 & 4 & 1 & 2 \\ 4 & 1 & 2 & 3 \end{vmatrix}.

determinantrow operationsa circulant determinant

Approach:

Use row operations to simplify the matrix — first make a common factor appear in the first row, and then expand along a column to reduce it to a 3×3 determinant and finish the calculation.

Steps:

  1. Add rows 2, 3 and 4 to row 1: R1→R1+R2+R3+R4R_1 \to R_1+R_2+R_3+R_4, so the first row becomes (10,10,10,10)(10, 10, 10, 10)
  2. Take out the common factor 10: D=10∣1111234134124123∣D = 10 \begin{vmatrix} 1 & 1 & 1 & 1 \\ 2 & 3 & 4 & 1 \\ 3 & 4 & 1 & 2 \\ 4 & 1 & 2 & 3 \end{vmatrix}
  3. Use the first row to clear the first column of the other rows: R2→R2−2R1R_2 \to R_2-2R_1, R3→R3−3R1R_3 \to R_3-3R_1, R4→R4−4R1R_4 \to R_4-4R_1 D=10∣1111012−101−2−10−3−2−1∣D = 10 \begin{vmatrix} 1 & 1 & 1 & 1 \\ 0 & 1 & 2 & -1 \\ 0 & 1 & -2 & -1 \\ 0 & -3 & -2 & -1 \end{vmatrix}
  4. Expand along the first column: D=10∣12−11−2−1−3−2−1∣D = 10 \begin{vmatrix} 1 & 2 & -1 \\ 1 & -2 & -1 \\ -3 & -2 & -1 \end{vmatrix}
  5. Compute the 3×3 determinant: =10[1((−2)(−1)−(−1)(−2))−2(1(−1)−(−1)(−3))+(−1)(1(−2)−(−2)(−3))]= 10[1((-2)(-1)-(-1)(-2)) - 2(1(-1)-(-1)(-3)) + (-1)(1(-2)-(-2)(-3))] =10[1(2−2)−2(−1−3)−1(−2−6)]= 10[1(2-2) - 2(-1-3) - 1(-2-6)] =10[0+8+8]=10×16=160= 10[0 + 8 + 8] = 10 \times 16 = 160

Answer: D=160D = 160

Problem 2

Let AA be a 3×3 matrix with ∣A∣=2|A| = 2. Find ∣2A∗−3A−1∣|2A^* - 3A^{-1}|.

determinantthe adjoint matrixthe inverse matrixproperties of a matrix

Approach:

First replace the adjoint matrix using A∗=∣A∣A−1A^* = |A|A^{-1}; the whole expression then simplifies to a multiple of A−1A^{-1}, and you finish with the relation between the determinant and the inverse.

Steps:

  1. By the property of the adjoint matrix: A∗=∣A∣A−1=2A−1A^* = |A|A^{-1} = 2A^{-1}
  2. 2A∗−3A−1=2(2A−1)−3A−1=4A−1−3A−1=A−12A^* - 3A^{-1} = 2(2A^{-1}) - 3A^{-1} = 4A^{-1} - 3A^{-1} = A^{-1}
  3. ∣A−1∣=1∣A∣=12|A^{-1}| = \frac{1}{|A|} = \frac{1}{2}

Answer: ∣2A∗−3A−1∣=12|2A^* - 3A^{-1}| = \frac{1}{2}

II. Matrices

Problem 3

Let A=(101020101)A = \begin{pmatrix} 1 & 0 & 1 \\ 0 & 2 & 0 \\ 1 & 0 & 1 \end{pmatrix}, and let the matrix XX satisfy AX+E=A2+XAX + E = A^2 + X. Find XX.

a matrix equationthe inverse matrixmatrix operations

Approach:

First move the terms to rearrange the matrix equation into (A−E)X=A2−E(A-E)X = A^2-E, then check whether A−EA-E is invertible and use A2−E=(A−E)(A+E)A^2-E = (A-E)(A+E) to simplify the computation of XX.

Steps:

  1. Rearrange the equation: AX−X=A2−EAX - X = A^2 - E, that is, (A−E)X=A2−E(A-E)X = A^2 - E
  2. A−E=(001010100)A - E = \begin{pmatrix} 0 & 0 & 1 \\ 0 & 1 & 0 \\ 1 & 0 & 0 \end{pmatrix}, and ∣A−E∣=−1≠0|A-E| = -1 \neq 0, so it is invertible
  3. A2−E=(A−E)(A+E)A^2 - E = (A-E)(A+E), so X=(A−E)−1(A−E)(A+E)=A+EX = (A-E)^{-1}(A-E)(A+E) = A+E
  4. X=A+E=(201030102)X = A + E = \begin{pmatrix} 2 & 0 & 1 \\ 0 & 3 & 0 \\ 1 & 0 & 2 \end{pmatrix}

Answer: X=(201030102)X = \begin{pmatrix} 2 & 0 & 1 \\ 0 & 3 & 0 \\ 1 & 0 & 2 \end{pmatrix}

Problem 4

Let AA be an n×nn\times n matrix satisfying A2−3A+2E=0A^2 - 3A + 2E = 0. Prove that AA is invertible and find A−1A^{-1}.

invertibilitythe inverse matrixa matrix polynomial

Approach:

Rearrange the given identity into a product of AA and another matrix equal to the identity matrix, and then read off the conclusion directly from the definition of the inverse.

Steps:

  1. From A2−3A+2E=0A^2 - 3A + 2E = 0 we get A2−3A=−2EA^2 - 3A = -2E
  2. That is, A(A−3E)=−2EA(A - 3E) = -2E; multiplying both sides by −12-\frac{1}{2}: A⋅3E−A2=EA \cdot \frac{3E-A}{2} = E
  3. By the definition of the inverse matrix, AA is invertible, and A−1=3E−A2A^{-1} = \frac{3E-A}{2}

Answer: AA is invertible, and A−1=3E−A2A^{-1} = \frac{3E-A}{2}

III. The Rank of a Matrix and Systems of Linear Equations

Problem 5

Find the rank of the matrix A=(12−1324−2636−391211)A = \begin{pmatrix} 1 & 2 & -1 & 3 \\ 2 & 4 & -2 & 6 \\ 3 & 6 & -3 & 9 \\ 1 & 2 & 1 & 1 \end{pmatrix}.

the rank of a matrixelementary row operationsrow echelon form

Approach:

Use elementary row operations to reduce the matrix to row echelon form; the number of non-zero rows in the row echelon form is the rank of the matrix.

Steps:

  1. Elementary row operations: R2→R2−2R1R_2 \to R_2-2R_1, R3→R3−3R1R_3 \to R_3-3R_1, R4→R4−R1R_4 \to R_4-R_1 (12−1300000000002−2)\begin{pmatrix} 1 & 2 & -1 & 3 \\ 0 & 0 & 0 & 0 \\ 0 & 0 & 0 & 0 \\ 0 & 0 & 2 & -2 \end{pmatrix}
  2. Swap rows: (12−13002−200000000)\begin{pmatrix} 1 & 2 & -1 & 3 \\ 0 & 0 & 2 & -2 \\ 0 & 0 & 0 & 0 \\ 0 & 0 & 0 & 0 \end{pmatrix}
  3. The row echelon form has 2 non-zero rows, so r(A)=2r(A) = 2

Answer: r(A)=2r(A) = 2

Problem 6

Discuss the solutions of the system {x1+x2+λx3=1x1+λx2+x3=1λx1+x2+x3=1\begin{cases} x_1 + x_2 + \lambda x_3 = 1 \\ x_1 + \lambda x_2 + x_3 = 1 \\ \lambda x_1 + x_2 + x_3 = 1 \end{cases}, and when there are infinitely many solutions find the general solution.

systems of linear equationsa discussion with a parameterCramer's rulethe general solution

Approach:

First compute the coefficient determinant as a factored expression in the parameter, and then split into cases according to whether the determinant is zero; when infinitely many solutions appear, find a fundamental system of solutions and the general solution separately.

Steps:

  1. The coefficient determinant: ∣A∣=∣11λ1λ1λ11∣=−(λ+2)(λ−1)2|A| = \begin{vmatrix} 1 & 1 & \lambda \\ 1 & \lambda & 1 \\ \lambda & 1 & 1 \end{vmatrix} = -(\lambda+2)(\lambda-1)^2
  2. When λ≠1\lambda \neq 1 and λ≠−2\lambda \neq -2, ∣A∣≠0|A| \neq 0 and the system has a unique solution.
  3. When λ=1\lambda = 1, the three equations coincide, all being x1+x2+x3=1x_1+x_2+x_3=1, and there are infinitely many solutions. The general solution: (1,0,0)T+k1(−1,1,0)T+k2(−1,0,1)T(1,0,0)^T + k_1(-1,1,0)^T + k_2(-1,0,1)^T (where k1,k2k_1,k_2 are arbitrary constants)
  4. When λ=−2\lambda = -2, row operations on the augmented matrix produce a contradictory row 0=30=3, so there is no solution.

Answer: for λ≠1,−2\lambda \neq 1,-2 there is a unique solution; for λ=1\lambda = 1 there are infinitely many solutions, with general solution (1,0,0)T+k1(−1,1,0)T+k2(−1,0,1)T(1,0,0)^T+k_1(-1,1,0)^T+k_2(-1,0,1)^T; for λ=−2\lambda = -2 there is no solution.

IV. Eigenvalues and Eigenvectors

Problem 7

Find the eigenvalues and eigenvectors of A=(2−1103−1213)A = \begin{pmatrix} 2 & -1 & 1 \\ 0 & 3 & -1 \\ 2 & 1 & 3 \end{pmatrix}.

eigenvalueseigenvectorsthe characteristic polynomial

Approach:

First write down the characteristic polynomial and solve it for all the eigenvalues, and then for each eigenvalue solve the homogeneous system to obtain all the corresponding eigenvectors.

Steps:

  1. The characteristic polynomial: ∣λE−A∣=∣λ−21−10λ−31−2−1λ−3∣=(λ−4)2(λ−2)|\lambda E - A| = \begin{vmatrix} \lambda-2 & 1 & -1 \\ 0 & \lambda-3 & 1 \\ -2 & -1 & \lambda-3 \end{vmatrix} = (\lambda-4)^2(\lambda-2)
  2. The eigenvalues: λ1=2\lambda_1 = 2, λ2=λ3=4\lambda_2 = \lambda_3 = 4
  3. For λ1=2\lambda_1 = 2: solve (2E−A)x=0(2E-A)x=0 to get the eigenvectors k1(−1,1,1)Tk_1(-1,1,1)^T (k1≠0k_1 \neq 0)
  4. For λ2=λ3=4\lambda_2 = \lambda_3 = 4: solve (4E−A)x=0(4E-A)x=0 to get the eigenvectors k2(1,0,2)T+k3(0,1,1)Tk_2(1,0,2)^T + k_3(0,1,1)^T (k2,k3k_2,k_3 not both zero)

Answer: the eigenvalues are 2,4,42, 4, 4; the corresponding eigenvectors are k1(−1,1,1)Tk_1(-1,1,1)^T and k2(1,0,2)T+k3(0,1,1)Tk_2(1,0,2)^T+k_3(0,1,1)^T.

Problem 8

Let AA be a 3×3 matrix with eigenvalues 1,2,31, 2, 3. Find ∣A∗+2A+E∣|A^* + 2A + E|.

eigenvaluesthe adjoint matrixdeterminantproperties of eigenvalues

Approach:

Use the fact that the eigenvalues of a sum of matrices are found by adding the eigenvalues pairwise, and those of a product by multiplying them pairwise, turning the determinant problem into the product of three numbers.

Steps:

  1. ∣A∣=1×2×3=6|A| = 1 \times 2 \times 3 = 6
  2. A∗=∣A∣A−1=6A−1A^* = |A|A^{-1} = 6A^{-1}, so the eigenvalues of A∗A^* are 6λ\frac{6}{\lambda}, that is, 6,3,26, 3, 2
  3. The eigenvalues of 2A2A are 2λ2\lambda, that is, 2,4,62, 4, 6
  4. The eigenvalues of EE are all 1
  5. The eigenvalues of A∗+2A+EA^* + 2A + E are 6+2+1=96+2+1=9, 3+4+1=83+4+1=8, 2+6+1=92+6+1=9
  6. ∣A∗+2A+E∣=9×8×9=648|A^* + 2A + E| = 9 \times 8 \times 9 = 648

Answer: ∣A∗+2A+E∣=648|A^* + 2A + E| = 648

V. Diagonalisation by Similarity and Quadratic Forms

Problem 9

Let A=(200032023)A = \begin{pmatrix} 2 & 0 & 0 \\ 0 & 3 & 2 \\ 0 & 2 & 3 \end{pmatrix}. Find the orthogonal matrix PP such that P−1APP^{-1}AP is diagonal.

orthogonal matrixdiagonalisation by similaritya real symmetric matrixGram-Schmidt orthogonalisation

Approach:

First find all the eigenvalues and the corresponding eigenvectors, then use Gram-Schmidt to normalise them and make them mutually orthogonal, and assemble them into the orthogonal matrix PP.

Steps:

  1. The characteristic polynomial: ∣λE−A∣=(λ−2)(λ−1)(λ−5)|\lambda E - A| = (\lambda-2)(\lambda-1)(\lambda-5), with eigenvalues λ1=1,λ2=2,λ3=5\lambda_1=1, \lambda_2=2, \lambda_3=5
  2. For λ1=1\lambda_1=1: the eigenvector ξ1=(0,−1,1)T\xi_1 = (0, -1, 1)^T, normalised p1=(0,−12,12)Tp_1 = (0, -\frac{1}{\sqrt{2}}, \frac{1}{\sqrt{2}})^T
  3. For λ2=2\lambda_2=2: the eigenvector ξ2=(1,0,0)T\xi_2 = (1, 0, 0)^T, normalised p2=(1,0,0)Tp_2 = (1, 0, 0)^T
  4. For λ3=5\lambda_3=5: the eigenvector ξ3=(0,1,1)T\xi_3 = (0, 1, 1)^T, normalised p3=(0,12,12)Tp_3 = (0, \frac{1}{\sqrt{2}}, \frac{1}{\sqrt{2}})^T
  5. The eigenvectors belonging to distinct eigenvalues of a real symmetric matrix are orthogonal, so P=(p1,p2,p3)=(010−1201212012)P = (p_1, p_2, p_3) = \begin{pmatrix} 0 & 1 & 0 \\ -\frac{1}{\sqrt{2}} & 0 & \frac{1}{\sqrt{2}} \\ \frac{1}{\sqrt{2}} & 0 & \frac{1}{\sqrt{2}} \end{pmatrix}
  6. P−1AP=diag(1,2,5)P^{-1}AP = \text{diag}(1, 2, 5)

Answer: P=(010−1201212012)P = \begin{pmatrix} 0 & 1 & 0 \\ -\frac{1}{\sqrt{2}} & 0 & \frac{1}{\sqrt{2}} \\ \frac{1}{\sqrt{2}} & 0 & \frac{1}{\sqrt{2}} \end{pmatrix}, and P−1AP=diag(1,2,5)P^{-1}AP = \text{diag}(1,2,5).

Problem 10

Decide whether the quadratic form f=2x12+5x22+5x32+4x1x2−4x1x3−8x2x3f = 2x_1^2 + 5x_2^2 + 5x_3^2 + 4x_1x_2 - 4x_1x_3 - 8x_2x_3 is positive definite.

quadratic formspositive definitenessleading principal minorseigenvalues

Approach:

First write the symmetric matrix of the quadratic form, then compute the leading principal minors one by one, and use the leading-principal-minor criterion to decide whether the quadratic form is positive definite.

Steps:

  1. The matrix of the quadratic form: A=(22−225−4−2−45)A = \begin{pmatrix} 2 & 2 & -2 \\ 2 & 5 & -4 \\ -2 & -4 & 5 \end{pmatrix}
  2. The leading principal minors:
    • Δ1=2>0\Delta_1 = 2 > 0
    • Δ2=∣2225∣=10−4=6>0\Delta_2 = \begin{vmatrix} 2 & 2 \\ 2 & 5 \end{vmatrix} = 10-4 = 6 > 0
    • Δ3=∣A∣=2(25−16)−2(10−8)+(−2)(−8+10)=18−4−4=10>0\Delta_3 = |A| = 2(25-16) - 2(10-8) + (-2)(-8+10) = 18 - 4 - 4 = 10 > 0
  3. All the leading principal minors are positive, so the quadratic form is positive definite.

(You can also check via the eigenvalues: they are 1,1,101, 1, 10, all positive, so it is positive definite.)

Answer: the quadratic form is positive definite.


Summary

This set of comprehensive exercises covers the core examination points of linear algebra:

  • Determinants: computing determinants, the determinants of the adjoint and inverse matrices
  • Matrices: matrix equations, inverse matrices, matrix polynomials
  • The rank of a matrix and systems of linear equations: finding the rank by elementary row operations, discussing the solutions of a system with a parameter
  • Eigenvalues and eigenvectors: computing eigenvalues, applying properties of eigenvalues
  • Diagonalisation by similarity: orthogonal diagonalisation of a real symmetric matrix
  • Quadratic forms: deciding positive definiteness (the leading-principal-minor method)

The three core problem types to master are computing determinants, the rank of a matrix and solving systems of linear equations, and eigenvalues and diagonalisation.

Found an issue with this content?

Tell us what could be improved.