This page focuses on the course 14.384 Time Series Analysis as it was taught by Prof. Anna Mikusheva in Fall 2013.
The course provides a survey of the theory and application of time series methods in econometrics. Topics covered will include univariate stationary and non-stationary models, vector autoregressions, frequency domain methods, models for estimation and inference in persistent time series, and structural breaks. The empirical applications in the course will be drawn primarily from macroeconomics.
This course is part of a sequence of courses on statistics and econometrics in the Economics Department.
Prof. Mikusheva strongly believes that the best way to learn the techniques of the course is by doing. Every problem set will include an applied task that may include computer programming. She does not restrict the students in their choice of computer language. The professor also does not require her students to write all programs by themselves from scratch. They may use user-written parts of codes found on the Internet, but the instructor does require that students understand the program used and properly document it with all needed citations of original sources. Collaboration with other students on problem sets is encouraged, however, the problem sets should be written independently.
Every fall semester
The students' grades were based on the following activities:
This course is an advanced topic class intended for Ph.D. students in economics and finance. It is also popular among students pursuing a Ph.D. in engineering.
During an average week, students were expected to spend 10.5 hours on the course, roughly divided as follows:
Met 2 times per week for 1.5 hours per session, for 26 sessions.
Students worked on five problems sets and a final exam taken at home.