Understand the concepts of stationarity, ergodicity, and the Wold decomposition theorem as theoretical foundations for time-series modelling. Identify, estimate, and select ARMA models using the Box-Jenkins methodology, information criteria, and residual diagnostics. Test for the presence of unit roots using the Augmented Dickey-Fuller, Phillips-Perron, and KPSS tests, and understand the consequences of non-stationarity for inference. Model cointegrating relationships using the Engle-Granger and Johansen procedures and specify error correction models. Specify and estimate Vector Autoregression (VAR) models and perform impulse response analysis and forecast error variance decomposition. Model conditional heteroscedasticity in financial time series using ARCH, GARCH, and extensions such as EGARCH and GJR-GARCH. Produce and evaluate forecasts using a range of models and benchmark them against simple alternatives using standard forecast accuracy metrics.

Time Series Analysis provides a comprehensive treatment of the statistical and econometric methods used to model, forecast, and interpret economic and financial data observed sequentially over time. The course covers both univariate and multivariate frameworks, progressing from classical ARMA models and spectral analysis to cointegration, vector autoregression, and GARCH-type volatility models. Emphasis is placed on the identification of non-stationarity, testing for unit roots and structural breaks, and the construction of reliable forecasts. Students will implement all methods using econometric software and critically evaluate the use of time-series techniques in macroeconomic and financial research.

Lecture-based teaching

Hamilton, J. D. (1994). Time Series Analysis. Princeton University Press. Enders, W. (2014). Applied Econometric Time Series (4th ed.). Wiley. Lütkepohl, H. (2005). New Introduction to Multiple Time Series Analysis. Springer.