The first part of this course covers nonlinear and non-Gaussian time series models, including models for discrete data with time-varying parameters and stochastic volatility models. We study estimation and inference methods, such as the extended Kalman filter and simulation-based methods.
The second part covers the formulation, estimation and testing of univariate and multivariate volatility models. We also discuss the use of high-frequency data in realized volatility measurement, and its use in volatility modelling.
The third part of the course covers non-linear regime-switching models, large-scale factor models, and forecast combination and evaluation.
For each topic, we discuss theoretical aspects of the models and methods. Real-data applications from economics and finance will show how the methods can be used in practice.
Selected chapters/pages in:
- Anders, Davis, Kreiss and Mikosch (eds.), “Handbook of Financial Time Series” (ADKM)
- Blasques (2019), “Advanced Econometric Methods” (B)
- Clements and Hendry (eds.), “Oxford Handbook of Economics Forecasting” (CH)
- Drukker (2011), “Missing Data Methods: Time-Series Methods and Applications”, vol. 27 (D)
- Elliott, Granger and Timmermann (eds.), “Handbook of Economic Forecasting”, vol. 1 (EGT)
- Franses and van Dijk (2000), “Nonlinear Time Series Models in Empirical Finance” (FD)
- Hamilton (1994), “Time Series Analysis” (H)
- Newey and McFadden (1994), “Large Sample Estimation and Hypothesis Testing” (NM)
- Potscher and Prucha (1997), “Dynamic Nonlinear Economic Models: Asymptotic Theory” (PP)
- Straumann (2005), “Estimation in Conditionally Heteroschedastic Time Series Models” (S)
- Van der Vaart (1998), “Asymptotic Statistics” (V)
- White (1996), “Estimation Inference and Specification Analysis”
Selected articles and working papers (W)