Time Series Analysis
Syllabus, Bachelor's level, 2ST093
- Code
- 2ST093
- Education cycle
- First cycle
- Main field(s) of study and in-depth level
- Statistics G1F
- Grading system
- Fail (U), Pass (G), Pass with distinction (VG)
- Finalised by
- The Faculty Board of Social Sciences, 31 May 2007
- Responsible department
- Department of Statistics
Entry requirements
Statistics, 30 credits points or equivalent
Learning outcomes
A student that has completed the course should
� have deeper knowledge of statistical theory and methods particularly common problems in economical social sciences especially economics.
� be able to estimate models for time-series data.
� be able to interpret the results of an implemented statistical analysis
� be aware of limitations and possible sources of errors in the analysis
� have ability to present results in oral and written form
Content
Overview of forecasting. Models for time series: Time-dependent seasonal components. Autoregressiva (AR), moving average (MA) and mixed ARMA-modeller. The Random Walk Model. Box-Jenkins methodology. Forecasts with ARIMA and VAR models.
Dynamic models with time-shifted explanatory variables. The Koyck transformation . �Partial adjustment� and �adaptive expectation� models. Granger's causality tests. Stationarity, unit roots and cointegration. Modelling of volatility: ARCH - and the GARCH-models.
Instruction
Lectures 6-10 hours a week.
Assessment
The examination comprises a written test at the end of the course and compulsory assignments, (laboratory sessions). Three grades are awarded for the course: not passed, passed, and passed with distinction.