PP-515: Advance Numerical Methods And Statistics
Tutorial-2 (Solutions of Linear Equations) Autumn-2022-2023
1. What are the advantages of iterative methods over direct methods for
solving a system of linear equations.
2. Apply factorization method to solve the following problem:
i. 4x1 + x2 + x3 = 4
x1 + 4x2 - 2x3 = 4
3x1 + 2x2 - 4x3 = 6 (Ans: x1 = 1, x2 = 1/2, x3 = -1/2)
ii. 3x + 2y +7z = 4
2x + 3y + z = 5
3x + 4y + z = 7 (Ans: x = 7/8, y = 8/3, z = -1/3)
iii. 2x + 3y + z = 9
x + 2y + 3z = 6
3x + y + 2z = 8 (Ans: x = 7/2, y = 1/2, z = 1/2)
3. Apply Gauss elimination method to solve the following problem:
i. x + 4y - z = -5
x + y - 6z = -12
3x - y - z = 4 (Ans: x = 1.64791, y = 1.14085, z = 2.08451)
ii. 3x + 4y - z = -6
-2y +10z = -8
4y - 2z = -2 (Ans: x = y = z = -1)
iii. -7x–3y +3z =12
2x +2y +2z =0
-x – 4y + 3z = -9 (Ans: x = -3, y = 3, z = 0)
4. Solve the equations using Gauss Jordan method.
i. x + 2y + 6z = 22
3x + 4y + z = 26
6x - y - z = 19 (Ans: x = 4, y = 3, z = 2)
ii. x + 3y + 2z + 5w = 11
−x + 2y − 2z + 5w = −6
2x + 6y + 4z + 7w = 19
5y + 2z + 6w = 5 (Ans: x = 5, y = -1, z = 2, w = 1)
iii. 2a + 8b – 4c = 0
2a + 11b + 5c = 9
4a + 18b + 3c = 11 (Ans: a = 2, b = 0, c =1)
5. Apply Jacobi Method to solve the following:
i. 10x1 + 2x2 + x3 = 9
2x1 + 20x2 - 2x3 = -44
-2x1 + 3x2 +10x3 = 22 (Ans: x1 =1, x2 = 2, x3 = 3)
ii. 83x1 +11x2 - 4x3 = 95
7x1 +52x2 + 13x3 =104
3x1 + 8x2+29x3 = 71 (Ans: x1 =1.06, x2 = 1.37, x3 =1.96)
iii. 4x1 + x2 + x3 = 2
x1 + 5x2 + 2x3 = -6
x1 + 2x2 + 3x3 = -4 (Ans: x1 = 0.9333, x2 = -1.0733, x3 = -1.100)
6. Apply Gauss- Seidel Method to solve the following:
i. 10x1 - 2x2 - x3 - x4 = 3
-2x1 +10x2 - x3 - x4 =15
-x1 -x2 +10x3 - 2x4 = 27
-x1 - x2 - 2x3 + 10x4 = -9
(Ans : x1 = 0.999, x2 = 1.999, x3 = 2.999, x4 = -0.001)
ii. 1.4x1 + 2.3x2 + 3.7x3 = 6.5
3.3x1+1.6x2 + 1.3x3 = 10.3
1.5x1+ 2.9x2 + 1.1x3 = 8.8 (Ans: x1 = 2.29, x2 = 1.97, x3 = -0.33)
7. Apply Gauss–Seidel Method to improve the solution correct upto 3 decimal
place of the system:
27x1 + 6x2 - x3 = 85
6x1 +15x2+2x3 = 72
x1+ x2+ 54x3 =11
starting with the solution x1 = 2.4, x2 = 3.5, x3 = 1.9.
8. Solve the following equation by Gauss-Seidel Method up to 3 iterations and
find the value of x. (correct upto 2 decimal place)
4x - 3y - z = 40
x - 6y + 2z = -28
x - 2y + 12z = -86 (Ans: x = 11.51)
9. The values of 𝑥1 = 𝑥2 = 1.000 are the solution to
1.133𝑥1 + 5.281𝑥2 = 6.414
24.14𝑥1 − 1.210𝑥2 = 22.93
(i) Use four-digit arithmetic and Gaussian elimination with trivial pivoting to
find a computed approximate solution to the system.
(Ans. 𝑥1 = 0.9956, 𝑥2 = 1.001)
(ii) Use four-digit arithmetic and Gaussian elimination with partial pivoting to
find a computed approximate solution to the system.
(iii) (Ans. 𝑥1 = 1.000, 𝑥2 = 1.000)
10. Solve the system of equations by the Jacobi method, beginning with the
initial vector (0,0,0). Compare the rate of convergence when Gauss-seidel
is used with the same starting vector.
2𝑥 − 6𝑦 + 𝑧 = 11
−5𝑥 + 𝑦 − 2𝑧 = −12
𝑥 + 2𝑦 + 7𝑧 = 20
11. Consider the following linear system:
5𝑥 + 3𝑦 = 6
4𝑥 − 2𝑦 = 8
Can either Jacobi or Gauss-Seidel iteration be used to solve this linear
system? Why?
(Hint: Matrix is not diagonally dominant)
12. Consider the following linear systems:
a) 4𝑥 − 𝑦 = 15; 𝑥 + 5𝑦 = 9
b) −𝑥 + 3𝑦 = 1; 6𝑥 − 2𝑦 = 2
Start with 𝑃0 = 0
(i) Use Jacobi iteration to find 𝑃𝑘 (k=1,2,3). Will Jacobi iteration
converge to the solution ?
(ii) Use Gauss-Seidel iteration to find 𝑃𝑘 (k=1,2,3). Will Gauss-Seidel
iteration converge to the solution ?