I'll be honest: I find this course difficult to describe. Will you learn about linear
regression? Of course, although you shouldn't be starting from ground zero. Will you
learn how to use R for linear regression? Of course, but that in itself isn't very ambitious;
after all, dozens of software packages provide tools for linear regression. I don't want
you to leave the course feeling that you have learned about a limited set of tools, allowing
you to do only certain types of analyses. I want you to feel prepared to face the
unexpected, equipped with a set of skills enabling you to adapt to the inevitable surprises
of data analysis. When faced with a fresh challenge, I want you to think, "I may not
know the answer, but I bet I can figure it out." Someday, I want you to think, "that was
one of the most practically useful courses I had at Yale."
Along the way, we'll learn about computer programming, algorithms, data structures, probability, statistics, simulation, numerical techniques, optimization, graphical methods, and computational efficiency. You should be able to think critically about data, use graphical and numerical summaries, apply standard statistical inference procedures (when appropriate) and draw conclusions from such analyses. But most importantly, you should be willing to break out of the box and conduct new, innovative analyses of problems when standard analyses may not be appropriate.
This course will be computationally intensive, and there is no substitute for getting your hands dirty. I expect to make my share of mistakes this semester (some intentional, some not), and we'll learn from them together. In data analysis, I believe you learn as much (and sometimes more) when things "don't work" than when they go as planned. You have succeeded when you can figure out why something doesn't work (or why some analysis isn't appropriate) and deduce an appropriate course of action as a result. You must be willing to try out new things and to make mistakes -- you can't break the computer (at least, it won't go up in smoke), and the sky won't fall. Seek to understand the mistakes, and move onward.
Course Syllabus (Fall 2009)
Department of Statistics
Jay's Home Page