The subject of this course is how to do econometrics and how to evaluate the econometric research of others. We will consider the selection and use of data to study a question, the design of an econometric model to fit a particular purpose, and the estimation and testing of that model. A large amount of lecture time will be devoted to formal presentation of commonly used estimation techniques, to which most students will have already been exposed in previous courses. However, unlike most courses in econometric theory, the presentation in the current course will be driven by a systematic investigation of specific research articles and extended examples from the econometrics literature. The usual approach will be to discuss (1) the economic, political, and policy issues that motivate the application, (2) the econometric techniques and issues related to the application, and (3) the results of the empirical analysis and what we can learn from them. For illustration, we will use STATA, the leading computer software for econometric analysis.
Some of the research questions that will be addressed are: Where should I begin, and how do I know if my problem is interesting enough? From where can I get data? Can I fit my regression by least squares, or do I need to use instrumental variables? Is it possible to use simulations in order to learn more about the performance of these estimators in my particular sample? What should I do if the dependent variable is binary? What if the data are trending? What if they have a panel structure? What kind of results should I include, and what are the conclusions that I can draw from them?