Modeling of Trends in Time Series based on Maximum Autocorrelation Factors

Warning

This publication doesn't include Faculty of Economics and Administration. It includes Faculty of Science. Official publication website can be found on muni.cz.
Authors

HÜBNEROVÁ Zuzana

Year of publication 2008
Type Article in Proceedings
Conference XVIII. letná škola biometriky BIOMETRICKÉ METÓDY A MODELY V PÔDOHOSPODÁRSKEJ VEDE, VÝSKUME A VÝUČBE
MU Faculty or unit

Faculty of Science

Citation
Field Applied statistics, operation research
Keywords Maximum autocorrelation factor; Trend; Time series; Ozone;
Description Description of trends in concurrent time series can be based on a nonparametric method suggested by P. Switzer [Analysis of Monitoring Data for Time Trends Using Maximum Autocorrelation Factors. International Association of Mathematical Geology, IAMG06, Liege. 2006]. In this approach, trends in individual time series are expressed as linear combinations of a chosen number of maximal autocorrelation factors an analogy of principal components focusing on the autocorrelation instead of the variance of a component. This paper describes basic ideas and chosen properties of the maximum autocorrelation factors analysis. An application of the considered method in analysis of trends of ozone time series at chosen European monitoring stations is also presented.
Related projects:

You are running an old browser version. We recommend updating your browser to its latest version.