Modules | Area | Type | Hours | Teacher(s) | |
TIME SERIES ECONOMETRICS | SECS-P/05 | LEZIONI | 42 |
|
Students are expected: - to know how to model, estimate and predict stationary and non-stationary time series, in a univariate and multivariate framework; - to understand and analyze the short and long run structure of the dynamic econometric models, acquiring knowledge of recent changes in the methodology of econometric analysis of time series; - to understand the importance of methodological changes in relation to the empirical modelling with macroeconomic and financial data.
Students are assumed to have had a previous course in Econometrics. A good grasp of basic mathematical statistics and linear algebra is necessary.
Suggested reading : The mathematical appendix in Hamilton gives a summary of useful mathematical and statistical tools.
Delivery: face to face
Learning activities: attending lectures
Attendance: Advised
Teaching methods: Lectures
UNIVARIATE TIME SERIES MODELS
MULTIVARIATE TIME SERIES MODELS
- Dynamic models with stationary variables (ADL model, Adaptive expectations, Partial adjustment)
- Models with non-stationary variables
- Vector autoregressive models
- Cointegration: the multivariate case
[Illustration: the expectations theory of the term structure, volatility in daily exchange rates, analysis of price/earnings ratio, long-run purchasing power parity, money demand and inflation.] Software:PcGive, Gretl and E-Views.
Lecture notes (available on https://elearning-old.ec.unipi.it/);
Verbeek M. (2012), A Guide to Modern Econometrics, John Wiley and Sons.
Juselius, K., The Cointegrated VAR Model, Methodology and Applications. Oxford University Press, 2007.
Hamilton James D., Time Series Analysis. Princeton University Press, 1994.
Methods: Final written exam
Time at disposal: 90 minutes (answer 4 or 5 questions). The student must demonstrate his/her knowledge of the course material and to organise an effective and correctly written reply.