Numerical Analysis 1, MATH545

7:00 - 8:15 Monday and Wednesday, Maybank 223
Dr. Brenton leMesurier

Office Hours: Mon, Tue, Wed., Fri 11am-noon and by arrangement (including evenings up till 7PM Mon.-Thu.)

Office Location: Room 103, 4 Greenway (between Maybank and Randolph Hall).

Phone: 953-5917

My Email address, obfuscated to avoid automated collection for junk mail purposes, is got by attaching my CofC username "lemesurierb" to the CofC host specifier "@cofc.edu".

Suggested References

Numerical Analysis: Mathematics of Scientific Computing by David Kincaid and Ward Cheney (3rd ed.),
Numerical Analysis by Richard Burden and J. Douglas Faires (7th ed.).
Earlier editions are satisfactory too.

News

Monday April 23
First versions of Project 3 are due.
Thursday April 12
The folder math545-project1 has been updated with a sample of how a complete project report could look.
Tuesday April 10
The description of Project 3 is now available. First versions are due by our last class, Monday April 23.
Tuesday March 27
As mentioned below, there are expanded collections of handouts and Matlab sample files, with the option of simple directory listings for these, handouts and matlab. This includes the second versions of function for the romberg method and a testing script for it.
Wednesday March 21
Assignment 3, the problems from Kincaid and Cheney already mentioned in class, is now available as a full printed descriptions for those of you who do not have K&C. It is due next Friday, March 30 but if handed in earlier next week I will give you worked solutions immediately.
Wednesday March 21
The notes from Monday's class on the handout on an Adaptive Runge-Kutta method is available, along with the Matlab sample files mentioned there and used in Monday's class.
Thursday March 15
A folder math545-project2 has the project handout and various Matlab files to use with it. Mostly you will be writing "run scripts" derived from the two such scripts there.

Course Overview

This course is a study of methods for computing accurate approximate numerical solutions to mathematical problems, typically arising as models of questions from science and engineering. The numerical solution of ordinary differential equations will be used as an organizing theme for the study of other topics: above all solving simultaneous equations by direct and iterative methods, and other topics from linear algebra such as computing eigenvalues and eigenvectors, approximation of functions, approximating derivatives, and definite integrals.

We will also address the practical issues of using software packages and designing and implementing computer programs to solve such problems. The main computational tool will be the MATLAB package for interactive computation and graphics as well as programming, which is provided on the Macintoshes in the computer classroom Maybank 200.

Related to this will be a focus on presenting the results of all such computational work in a coherent form as would be expected in any workplace: explaining the choice of numerical methods on such bases as accuracy, robustness and efficiency; presenting the numerical results themselves with suitable tables and graphs; and discussing the meaning, reliability and accuracy of these results.

Assignments, handouts and other resources

Only the most recent and important are listed here, will more complete lists available at the following links for handouts and Matlab sample files. Or you can get a concise listings for downloads in folders handouts and matlab. From handouts you can also download the LaTeX source files (suffix .tex) for most handouts, as examples for those learning LaTeX.
Dr. Brenton leMesurier, Department of Mathematics, College of Charleston
Last revised Thursday April 12, 2007.