This course is concerned with some topics in modern time series econometrics. Its coverage begins with some of the fundamental concepts used to analyse stationary time series, before proceeding to the analysis of nonstationary (integrated) processes that have dominated recent research in theoretical and applied time series econometrics. The emphasis throughout is on maximum likelihood estimation of linear models, and both univariate and multivariate processes and models are examined. The course concludes with a treatment of continuous time models and models with nonlinearities in mean and variance.
Feedback for this module will occur through: class meetings, where we will go over the answers to problem sets and where you will be able to ask questions about your own method of solution; outline answers that will be posted on the website for the module that will give you written guidance on the appropriate method to approach the problem sets and tests; and office hours, where any additional questions can be addressed. You should ensure that you use these methods to understand how to improve your own performance.
Upon successful completion of this course students will have acquired an appreciation of econometric methods applicable to the analysis of models for economic time series, covering stationary and nonstationary situations in both univariate and multivariate contexts. They should understand the methods of estimation and inference as applied in these models, be able to derive the properties of some econometric methods applicable to time series and be prepared for the use of these methods in their own empirical research.
- Module Supervisor: Marcus Chambers