Research subject Time series analysis
When measuring a variable periodically in time, the observations form a time series. Unlike many other areas in statistics, time series analysis is no more than around 100 years old.
Time series analysis is based on the type of data where a variable is regularly measured in time. The method is primarily used to decompose time series. For example, seasonal adjustment, identify and model the systematic variation, identify and model the time-based dependencies and forecasts.
Nowadays, the so-called Box-Jenkins models are perhaps the most commonly used and many techniques used for forecasting and seasonal adjustment can be traced back to these models.
Another line of development are non-linear generalizations, mainly ARCH (AutoRegressive Conditional Heteroscedasticity) - and GARCH- (G = Generalized) models which have proved very useful, especially for financial time series. The invention of them and the release of a way to correct the models for errors, provided C. W. J. Granger and R. F. Engle with the Nobel Memorial Prize in Economic Sciences in 2003.
Related research subject
Statistics![Foto: Håkan Slättman Allé Humlegården](/polopoly_fs/1.612762.1652792281!/image/image.jpg_gen/derivatives/landscape_690/image.jpg)
On this page
Researchers
Andriy Andreev
Universitetslektor
![](/webb2021/img/fallback_image_profile.png)
Oskar Gustafsson
Universitetslektor
![](/webb2021/img/fallback_image_profile.png)
Oscar Oelrich
Universitetslektor
![](/webb2021/img/fallback_image_profile.png)
Pär Gunnar Victor Stockhammar
Universitetslektor
![](/webb2021/img/fallback_image_profile.png)
Mattias Villani
Professor
![MattiasPhoto](/polopoly_fs/1.538032.1611756463!/image/image.jpg_gen/derivatives/box_260/image.jpg)