This course offers a modern introduction to the broad and dynamic field of time series analysis, with a focus on applications in macroeconometrics and financial econometrics. It balances theoretical foundations with practical tools and applications, providing students with a comprehensive understanding of key methodologies.
In the first part of the course, we focus on the most fundamental class of time series models: the ARMA model. Students will learn how to specify, estimate, diagnose, and forecast using ARMA models. We also address the crucial topic of stationarity, including methods for handling trends, seasonality, and unit roots.
The second part of the course expands to multivariate time series models, particularly the VARMA framework. We explore cointegration analysis and modern approaches to estimating Structural VARs and local projections. The course also introduces State Space Models, covering their estimation via the Kalman filter and modern Bayesian filtering and smoothing techniques.
We then turn to the challenges of high-dimensional time series modeling, comparing sparse models (regularization techniques) with dense models such as dynamic factor models. Finally, we delve into key models used in financial econometrics, including univariate GARCH models and multivariate copula-based models. |
