SF2945 Time Series Analysis 6.0 credits
The overall purpose of the course is that the student should be well acquainted with basic concepts, theory, models and solution methods in time series analysis, models for "dependent" data. Such data are common in economical (e.g., the price development of a product) and in natural science (e.g., meteorological observations) applications.
Educational levelSecond cycle
Academic level (A-D)D
Grade scaleA, B, C, D, E, FX, F
At present this course is not scheduled to be offered.
To pass the course, the student should be able to do the following:
- identify trends and seasonal variations in time series
- define and calculate expectation, covariance function and spectral distribution and analyse their relations
- estimate the above mentioned quantities for time series data and quantify the uncertainty in these estimates
- predict on real time series of different lengths, for instance by recursive methods
- define and apply parametric models of ARMA type and analyse the properties of the models
- fit ARMA models to real data
- explain the generalisations ARIMA and FARIMA of ARMA models
- analyse data with parametric variance models of ARCH type
- describe in a broad sense Kalman filters
To receive the highest grade, the student should in addition be able to do the following:
- Combine all the concepts and methods mentioned above in order to solve more complex problems.
Course main content
General introduction to time series. Stationary and non-stationary models, e.g. ARMA- and ARIMA-models. Prediction of time series. Spectral theory. Estimation of parameters and of the spectra. Filtering.
Previous knowledge is assumed equivalent to SF1906 (5B1506).
Brockwell and Davis: Introduction to Time Series and Forecasting, Springer-Verlag.
- TEN1 - Examination, 4.5 credits, grade scale: A, B, C, D, E, FX, F
- ÖVN1 - Assignments, 1.5 credits, grade scale: P, F
Requirements for final grade
Replaced by SF2943 from 11/12.
Course plan valid from:
Examination information valid from: Autumn 07.