Course name: Numerical Laboratory  

Second semester 2006

Leader: Prof D P Laurie, Van der Sterrgebou 2025, dpl@sun.ac.za

Time and place: Open to negotiation

Course content:

  We try to get answers to the following questions:
  1. How do we compute numbers for which there is no closed formula?
  2. How do we know how many digits of our answers can be trusted?
  3. What can we do to improve the accuracy of our answers?

Prerequisites:

1.  Any undergraduate course in computational mathematics, approximation
theory or numerical analysis, eg Mathematics 354 or Applied Mathematics 324,
or the AIMS numerical analysis module, or a first semester honours course 
on such a subject.  
2.  Expertise in one or more of Matlab, Octave, Pari-GP,
Maple, Mathematica, C++ or Fortran.
3.  An inquiring temperament and a delight in problem solving.

Contra-indications:

Students who like to know this is the syllabus, this is the textbook,
on this date we do this, for the exam you learn that, are advised to
select some other course.

Method of presentation:

The course has no formal theory component.  Problems are distributed
to the students, who have one or more weeks (depending on the degree
of difficulty of the problem) in which to do the problems.
Students must find their own study material in the library or on the
internet.

A discussion session is held each week, during which the students give
feedback on their progress with the problems.  As far as possible students
are expected to learn from each other in a teamwork situation.  The leader
will only intervene in cases where students despite serious efforts
are getting frustrated.  He will then, depending on the requirements
that arise, give hints or references, or present brief lectures on the
necessary topics.

Evaluation:

Continual.