• describe and apply various models for time series data,
• estimate and interpret parameters in these models,
• compute forecasts and evaluate forecasting methods,
• describe and apply structural and causal interpretations of time series models,
• report the results of the above clearly and with correct use of technical terminology.

A time series consists of observed data which can be ordered according to time of observation. Time series data is used in many disciplines, among these economics and finance. This course will introduce classical and modern methods for time series analysis.

In the first part of the course, we will introduce AR-processes, ARMA-processes, ARCH- and GARCH-processes as well as their multivariate counterparts. In addition, we will study estimation and forecasting in time series models.

In the second part of the course, we will study structural and causal interpretations of time series models, and we will introduce SVAR-processes. We will describe time series models as causal models, and we will define the notion of a causal effect in a time series context. We will also discuss partially observed time series.

The course consists of theoretical and practical components. We will use R for practical examples.

• Classic and basic theory
• New theory
• Teacher’s own research
• Methodology
• Models

• Discussion, critical reflection, modelling