Math 240 Section E1 Syllabus
Scientific Computation
Winter 2019 (#3830)

Course content outline:

Learn to create stylish, convincing, easy-to-read, documented, illustrated reports using MATLAB.

MATLAB programming including arithmetic, relational, and logical operators, file input and output, script and function files, iterative (for and while) and conditional (if-elseif-else) flow control and plotting

Finding roots of equations and solving systems of linear algebraic equations

Curve fitting and interpolation

Several types of optimization

I will add some topics regarding statistical analysis of data

Numerical differentiation and numerical integration

Numerical solution of ordinary differential equations including initial value problems and boundary value problems

Eigenvalue and eigenvector problems and the SVD

Spectral analysis (fast Fourier transform) and other topics, such as Partial Differential Equations, as time and student interest allow.

Homework and take-home tests are due on at the start of class on the first class day of the week starting in week 3. If you anticipate an absence, turning the assignment in early (or emailed pdf attachment BEFORE class time) is always an option.

In the event of a school weather closure, test and homework due dates are moved to the next open school day.

You are responsible for showing up daily, participating in class discussions and contributing to in-class problem solving, and keeping current with all class events. At least once a week, but particularly on days when I observe sparse attendance, I will take roll and this record will constitute the "class participation" component, one sixth of the points you earn for your grade. Don't expect an A (and it is unlikely you will get a B) in this class if you are "unlucky" with these participation spot-checks.

Course materials will posted at a near-daily rate at:

You must have access to working installations of certain software to take this class. The commercial software MATLAB is required. The open source software GNU Octave is an unsupported (by me) command-line-based substitute for MATLAB in some cases.

MATLAB toolboxes are more extensive (for applications well beyond the scope of this course) and MATLAB has, of course, the polish of commercial software. MATLAB is the software of choice in the Engineering courses at many schools including UW. I highly recommend it. For our applications, Octave commands can be chosen to be identical to equivalent MATLAB commands. (You must make this choice so I can verify your solutions if you use Octave.)

You must acquire the software listed above and create a working installation on a computer to which you have easy access (you will be spending MANY hours creating and debugging software solutions!) and this must be done during the first few days of class. You won't be able to do any homework or understand what we are doing without it.

I hope that you are actually able to do many of the things that are taught in the prerequisite class. That means Math 208, linear algebra. It would be great if you know about linear transformations on the vector space of real ordered n-tuples (such as ordered pairs and ordered triples) and how they are related to matrices, and what THAT has to do with solving n linear equations in m unknowns. The basic algebra of matrices will be used almost daily. The words: dot product, orthogonal projection, eigenvalue, eigenvector, basis, dimension, dependence and independence, span and subspace, linear transformations and the matrices of these with respect to bases will be used in the course, as will row reduced echelon form.

Linear Algebra itself has prerequisites here at BC: you need to have taken Calculus to sign up for it, and I hope you do remember some Calculus.

We will develop the necessary ideas about Differential Equations as they pop up, though folks who have had a course in Differential Equations will find that very useful here.

Much of the class time will be spent talking about the mathematical ideas behind the software tools you are using in the problems you are doing, and understanding what you are doing is not optional. This understanding will be tested through your ability to solve the assigned problems. It is this aspect of our course that makes it appropriate to teach this class through the math program: it is not a purely "cut and paste in the dark to get an answer" class, nor is it a programming class, though we will definitely want to get answers to practical problems and we will learn some MATLAB programming.

Here is how I see all these things fitting together. We will discuss some scenario that comes up in applications and learn a little vocabulary for that subject, outlining but not justifying any physical or other principles which have been found to be important for that application. (After all, we must leave something for your physics and engineering and economics classes.) We will discuss one or more mathematical techniques used to deal with important questions in this subject area. We will learn one or more MATLAB methods to input any data required and do the indicated calculation. We will discuss error issues. We will learn how to present a conclusion using MATLAB tools, including file output and visualization tools. In the HW and Tests you will have an opportunity to put all these skills together to create small "reports", one for each problem.

Generally, if a choice must be made we will trend toward the "simpler or easier to understand" rather than the "sophisticated and quicker." My point of view here is that if you understand the basic issues you can embellish later.

Topics will be chosen according to intrinsic interest (always in the eye of the beholder) but also because they invoke, in their solution, a spectrum of MATLAB talents.

MATLAB is a vast program, with a near infinitude of options and packages and toolkits, and this is true of competitor programs such as Maple and Mathematica too. We will only scratch the surface of techniques you may come to learn in some special area or another. Be aware that there will be MANY methods and variants and alternatives leading to solutions.

But we hope to make a good beginning with MATLAB, and also introduce you to ideas of modeling and error analysis and scientific calculation, and their limitations, and report-writing that will support your later work in your field with any of the common computer-aided-mathematics programs you may see.

The complete textbook corresponds to a very challenging course, and I can't imagine that anyone does all the topics included in the text in 11 (or even 16) weeks. We will, however, select topics from most chapters, and cover more than half the contents of the book.

Important Dates :

Grades will be based on 4 Tests (40 points each), the Final (40 points) and 5 Homework Assignments (16 points each, 80 points total) and a Participation component (40 points).

I will select your best 3 out of the 5 major tests and add to that your Homework and Participation points to compute your grade.

Participation points are earned, essentially, by being there when I record them and are intended to provide incentive for daily on-time attendance. If you are not there at that moment, which will be at random moments during the class time about twice a week, you will not get these points. Fair warning: since they account for about 17% of the grade you will not get an A if you miss many days or are regularly late. On the other side of that, scrupulous attendance can really help you.

I expect you to do your own work on take-home tests, but on homework I will not object to collaboration. I do understand that online resources are readily available. However you are responsible for personally writing up these solutions, and for understanding everything on any paper you turn in. If you don't understand your friend's tip, you can't use it.

The physical copy of take-home tests or homework items submitted for grading MUST be CAREFULLY ORGANIZED, PROBLEMS PRESENTED IN ORDER, STAPLED BY THE START OF CLASS TIME with a LEGIBLY, COMPLETELY and ACCURATELY FILLED IN COVER SHEET which can be found at

Each problem within an assignment is to be presented as its own little "stand-alone" report starting on its own page. The idea here is that a person knowledgeable in the field with which that problem deals and who understands how MATLAB works should be able to sit down and comprehend what the problem is, what you have done to solve it and, given what you have written, have a basis to agree or not with your conclusion. That conclusion should be unambiguously identified and sometimes discussion of error (or even validity of the method used) will be warranted. A list of answers is absolutely not what I am looking for here.

You turn in two copies of each test or assignment by the start of class time on the due date. One will be an electronic copy. The other will be an IDENTICAL physical copy which I will grade and return to you.

The electronic copy should be sent to me by email attachment with your work all in one file in pdf or word format with a name that will clearly identify it as yours among the hundreds of other similar files I will get over this quarter. JonesTest3.pdf is good. BillyMatlab.pdf is bad. This file should contain as first page the Cover sheet (or a decent simulacra) properly filled out, found in the link above. If you are going to miss class on a due date you can turn in your identical hardcopy the next day, but the e-copy (whatever you have finished) must be on time.

Grade lines will 50%-F to D, 60%-D to C, 80%-B to C and 90%-B to A. (These intervals are divided into 3 equal subintervals for the +/- grade determination.)

Essential elements of class etiquette involve being respectful of your colleagues and the general classroom environment. They include the following.

Cell phones are disturbing, and people are distracted not just at the moment one goes off but for several minutes afterwards too.

Having personal conversations in class time distracts the people around you, even if is on a mathematical topic. If you have a question, catch my attention and ask it. Most likely you are not the only one with that question.

Walking in and out of class during class is absolutely not OK. It is at least as distracting as a cell phone.

Cheating is a poor way to try to pass a course and the faculty here at BC has agreed to be tough about instances of cheating and to announce this in our syllabi. If caught, at the least cheaters will receive an irrevocable 0 on the test in question and other administrative action is possible that may, for example, affect continued student status. Don't do it.