Aims:
The goal of this course is to provide students with an understanding and working knowledge of statistical techniques for the empirical analysis and forecasting of time series in macroeconomics and, to a lesser extent, finance. Although the focus of the course is primarily applied, there will also be some emphasis on the theoretical foundations of the techniques analyzed
Course outline:
1. Univariate Time Series
Topics include: Tests for structural breaks; Nonlinear models of the conditional mean; Models of the conditional variance: ARCH/GARCH
2. Multivariate Time Series
Topics include: VAR; Structural VAR; Impulse-response analysis; Granger causality; Cointegration
3. Elements of forecasting
Topics include: Forecasting with regression models; Model selection and information criteria; Forecast evaluation and Combination; Forecasting with many predictors: data-reduction methods
Module video presented by Raffaella Giacomini
Taught by: | Raffaella Giacomini and Saleem Bahaj |
Assessment: | 2 hours of lectures per week and weekly tutorial/practical classes with written assignments. The course will be examined by a 2-hour unseen written exam in Term 3. |
Suitable for: | Graduate students |
Prerequisites: | Successfully completed the Econometrics module in Term 1 and Enrolled on the UCL MSc Economics or UCL MSc Data Science & Public Policy (Economics route). |
Moodle page: | ECON0058 |