Numerical Methods for Statistics (Math3061)
Course Guide Book
Course Titles/Code: Numerical Methods for Statistics (Math3061)
Credit: 5EtCTS
Credit hour: 3hrs (3hrs lecture+1tutorial+1lab)
Module title/code: Numerical Methods for Statistics (Math-M3063)
Course Type: Supportive
Academic Year: 2023/24
Semester: one
Study Program: Undergraduate
Instructor’s Name: Solomon R.
Student’s Work load
Lecture Tutorial Assessment Lab Home Study Total
48 32 15 - 40 135
Course description This course covers basic concepts in error estimation solutions of non-linear
equations, solutions of system of linear equations and non-linear equations,
finite differences, numerical interpolations, numerical differentiation and
integration, introductory first order initial value problems (IVP), Taylor’s
method and Euler’s method.
The course intends to introduce students with finding numerical solutions to
statistical problems for each analytical solution can easily be obtained. It also
Objectives
aims to help students develop programming skills.
1
Learning outcomes
After taking this course, the student should be able to :
understand concepts of errors
understand a range of iterative methods for solving linear and non-linear systems of equations
understand the roles of finite differences
understand the concept of IVP
grasp the practical knowledge of interpolation in numerical solving of IVP
develop skills in translating numerical algorithms into computer programming
Course Outline
1. Basic concepts in error estimation (4 lecture hours)
1.1. Sources of errors
1.2. Approximation of errors
1.3. Rounding off errors
1.4. Absolute and relative errors
1.5. Propagation of errors
1.6. Instability
2. Nonlinear Equations (9 lecture hours)
1.1. Locating Roots
1.2. Bisection Method
1.3. Interpolation and Secant methods
1.4. Iteration Method
1.5. Conditions for convergence
1.6. Newton Raphson’s Method
3. System of equations (10 lecture hours)
3.1. Direct Methods for system of linear equations
2
3.1.1. Gauss's Method
3.1.2. Gauss's Method with Partial Pivoting
3.1.3. Jordan's Method
3.1.4. Matrix inversion using Jordan's Method
3.1.5. Matrix Decomposition
3.1.6. Tri Diagonal Matrix Method
3.2. Indirect Methods for system of linear equations
3.2.1. Gauss Jacobi Method
3.2.2. Gauss Seidel Method
3.3. Solving Systems of non-linear equations using Newton's method.
4. Finite differences ( 4 hours)
4.1. Shift operators
4.2. Forward difference operators
4.3. Backward difference operators
5. Interpolation (6 lecture hours)
5.1. Linear Interpolation
5.2. nth degree Interpolation
5.3. Lagrange's Interpolation formula
5.4. Newton Interpolation Formula (forward and backward difference formulas)
5.5. Application of interpolation
5.5.1. Differentiation
5.5.2. Integration (Trapezoidal & Simpson’s Rules)
6. First Order Initial Value Problems (9 lecture hours)
6.1. Definition of ODE and examples
6.2. Order of a differential equation, linear and nonlinear ODE
6.3. Nature of Solutions of ODE: particular and general solutions
6.4. Initial value problem
6.5. Existence of a unique solution (Picards theorem)
3
6.6. Method of separable of variables
6.7. Homogeneous equations
6.8. Exact equations, non-exact equations and integrating factor
6.9. Linear equations
6.10. Orthogonal trajectories
7. Numerical Methods for Initial Value Problems (6 lecture hours)
7.1. Taylor’s method of order n
7.2. Euler’s Method
7.3. Euler’s modified method
Teaching and Learning Methods
Lecture, Tutorials and Computer lab.
Modes of Assessment
Continuous Assessment 50%
Final Exam 50%
Total 100%
Method Assignment Test Quiz/Laboratory Final
Percent 20 20 10 50
Frequency 2 2 At least 2 1
