If you are thinking about taking Stat 434 in the Fall of 2007, I have a few tips and some practical information.
Statistics 434 is NOT a course for everyone. You really must have understood your earlier statistics courses --- in addition to having done well in them.
Also, you simply MUST be interested in financial time series and in the behavior of asset returns.
This is not a general course in time series, and, if you want to understand the uses of time series in medicine or acoustics or --- tons of other places --- well, this course would be a poor use of your time.
For such applications, you need methods that we do not cover. Moreover, many of the methods we do cover are only of interest for financial series. These include GARCH, Co-Integration, Tests for Stationarity, and many other techniques specific to financial time series.
My simple rule is this: If you are concerned if this is the right class for you --- It almost certainly is NOT.
The people who enjoy this class have experience, have confidence, have independent views, are articulate, are energetic, like to sort out problems on their own (or with a team member), and --- to an impressive extent --- have their life under control.
Once you have decided that you want to stake your claim among the committed few, then we can talk about technical stuff. In particular, I have debated back and forth about whether to use S-Plus (the tool we have used before) or R, the free generic statistical software.
The basic capabilities and syntax of R and S are equivalent, but for the moment, the financial tools in R are inadequate.
Accordingly, we are sticking with S-Plus. Still, within a year or two, you will probably be able to do with the (free) R what we will be doing with the (pricy) S-Plus.
In the Fall of 2007 we will again use the text by Zivot and Wang.
I came to this decision, only after much soul searching.
We are really only going to use about a fourth of its many pages, but there is no (legal) way that I could think of getting you just those pages that we need.
I wish that the book were (1) smaller (2) more focused on what we use (3) more "opinionated" about what works --- or doesn't (4) more generous with coaching about S-Plus and (5) more sincere in its engagement of modeling issues.
Eventually, I will write my own text for 434, but there is no chance that a workable version will be available in 2007, even 2008 is a long shot. So, we will have to make do. Naturally, as individuals, you may decide as just buy one copy per team, or you might make appropriate the xerox machine. What I can't do, though I am tempted, is to just copy out the pages that I think you need and make them available as a "bulk pack." What I can do is create a page (TBA) where I list the parts of Z&W that we will be using.
Still, when all is said and done, Z&W does contain information that we have to have, and the easiest way to get that information is just to bite the bullet and buy a copy.
Bring integrity!
Honest data analysis requires --- first of all --- that one is honest. This is harder than it sounds. Many published analyses fail to be genuinely honest.
We are all amazingly prone to various biases --- such as confirmation bias. We can't entirely escape the trap that these biases set for us, and it takes guts even to try.
The intention is to make Stat 434 as useful, as informative, and as enjoyable as any course that you have ever taken. It should be --- and can be --- a capstone experience.
This is totally possible if we manage to assemble a group of students who have
- a solid preparation,
- a genuine well-informed interest in the material, and
- a mature understanding of what it means to be intellectually honest.
We've been very lucky for three years running, but to keep being lucky requires a lot of thoughtful self-selection on the part of the students who commit to this course.
Back to the Stat 434 Home Page