Open course ware

I actually would like to take this occasion to praise and thank the TAs. We've been teaching this course for about a dozen years and we've had some really excellent groups of TAs, but this year's crop is really off scale, really outstanding. And we're all very grateful. [APPLAUSE] You know their names. I won't go through them all. But the fact of the matter is these TAs teach this course because they're told to teach the course, and at the same time they're doing all their thesis research, so they end spending about 168 hours a week on various kinds of work. So it's not a natural thing for them to spend an enormous amount of time, as they have been this year, really just off scale extraordinarily good. Here they are, Winston, Susan, Michelle, Sara, Divia, Jim, Sydney, Yasmine and Cha. So thank you all. At the same time, I'd also like to thank Claudette Gardel who runs this thing. This is a l...

Actually, the title is interesting. It is Materials of Equilibrium. A lot of people who look at the content would think of this as, you know, this comes out of a course that used to be called Thermodynamics of Materials which is a standard staple of students who would do graduate materials education. Almost everybody would have this as a requirement. We deliberately changed the title to Materials of Equilibrium to make it something broader. We really wanted it to be Standard Classical Thermodynamics, which is what we do, but wanted to broaden that into sort of modern metrics to understand and study the energetics of the materials. What we do is we essentially teach, first of all, we teach three large segments. One is classical thermodynamics. And the idea is that people need to see the general frame in which you can set up an energy description for materials. How they react with their environment. How they react with other systems. So we wanted to teach them first the gen...

Now, we're ready to actually start building models. So as usual, the first thing we're going to do is split our data into a training and a testing set. And for this problem, we're actually going to train on data from the 2004 and 2008 elections, and we're going to test on data from the 2012 presidential election. So to do that, we'll create a data frame called Train, using the subset function that breaks down the original polling data frame and only stores the observations when either the Year was 2004 or when the Year was 2008. And to obtain the testing set, we're going to use subset to create a data frame called Test that saves the observations in polling where the year was 2012. So now that we've broken it down into a training and a testing set, we want to understand the prediction of our baseline model against which we want to compare a later logistic regression model. So to do that, we'll look at the breakdown of the dependent variable...