MTH208 NUMERICAL ANALYSIS THE SECOND SEMESTER 2021/2022 RESIT EXAMINATION
Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit
MTH208
THE SECOND SEMESTER 2021/2022 RESIT EXAMINATION
BACHELOR DEGREE - Year 1
TIME ALLOWED: 2.5 hours
NUMERICAL ANALYSIS
Question 1 (5 marks).
show that the bisection method on an interval [a, b] gives a sequence with an error bound that converges at a linear rate to O .
Question 2 (15 marks).
(2-a) (5 marks). show that when Newton,s method is applied to the equation ①2 - 2 = O, the resulting iteration function is g(①) = ① + 从(2) ).
(2-b) (5 marks). Apply the developed Newton,s method to ind c4 using co = 1. (keep at least 9 decimal places in calculations)
(2-c) (5 marks). As an approximation to(2, how many correct decimal places does c4 have ?
Question 3 (15 marks).
some exact values of ^从 are tabulated here:
(3-a) (5 marks). Construct a degree-2 Lagrange interpolating polynomial P (从) that interpolates these data.
(3-b) (5 marks). use your interpolant from (a) to approximate ^3. Give a theoretical upper bound on the absolute error in this approximation. show that the actual absolute error is less than your upper bound.
(3-c) (5 marks). If P (从) is used to approximate^从 for 从 G [1, 4], the absolute error will depend on 从. Give a theoretical upper bound (independent of 从) on the absolute error.
Question 4 (1o marks).
The Trapezoidal rule applied to102 f (从)d从 gives the value 4 , and Simpson,s rule gives the value 2. Find the value f (1).
Question 5 (1o marks).
Find the constants c0 , c1 and 从1 such that the quadrature formula
l03 f (从)d从 = c0 f (o) + c1 f (从1 )
is exact for polynomials of as high a degree as possible.
Question 6 (1O marks).
(6-a) (5 marks). use Euler,s method with h = O.5 for the IVP
g、= g - t2 + 1
for t E [O)2] with initial value g(O) = O.5.
(6-b) (5 marks). Compute the theoretical largest value of h to ensure that |gi -wi | 参 O.5 (the answer should be in the form x.xxxx).
Question 7 (15 marks).
(7-a) (5 marks). Consider the following matrix A, whose LU factorization we wish to compute using Gaussian elimination:
l 4 A = ' 6 [ O
-8
5
-1O
1 」
7 '
-3 l
what will be the value of the initial pivot element if (no explanation required)
- No pivoting is used?
- Partial pivoting is used?
- Full pivoting is used?
(7-b) (5 marks). state one deining property of a singular square matrix A. suppose that the linear system A从 = b has two distinct solutions 从 and g. use the property you gave to prove that A must be singular.
(7-c) (5 marks). Describe an advantage of the Gauss-seidel algorithm over the Jacobi algorithm and one disadvantage.
Question 8 (2O marks).
Let A = Rm 根m . Recall that we say that A is symmetric positive deinite if A is symmetric, and vT Av 持 O for every nonzero vector v = Rm 根1 .
(8-a) (5 marks). suppose that A = X TX for some nonsingular square matrix X . Prove that A is symmetric positive deinite.
(8-b)(5 marks). suppose A has cholesky decomposition A = RTR where R is an upper triangular matrix,
and also A =R(教)TDR(教), where D is a diagonal matrix with positive diagonal entries. show that R = ^D R(教) .
(8-c) (5 marks). compute the cholesky factorization of the matrix A = ' O 4 [ 1 2
1 」
2 ' .
3 l
l 1 O 1 」
(8-d) (5 marks). For the matrix A = ' O 4 2 ' , compute an approximation to the largest eigenvalue by
[ 1 2 3 l
performing two steps of the power method with ①0 = [1, 1, 1]T .
2023-07-21