Open AccessA Hybrid Approach to Regime Shift Detection
Author(s) -
Alexander von Eye,
Wolfgang Wiedermann,
Stefan Weber
Publication year - 2019
Publication title -
journal for person-oriented research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.23
H-Index - 3
eISSN - 2003-0177
pISSN - 2002-0244
DOI - 10.17505/jpor.2019.04
Subject(s) - series (stratigraphy) , mean shift , a priori and a posteriori , computer science , point (geometry) , process (computing) , algorithm , point process , multivariate statistics , frequency analysis , data point , data mining , statistics , mathematics , pattern recognition (psychology) , artificial intelligence , machine learning , paleontology , philosophy , geometry , epistemology , biology , operating system
In this article, we propose a method for the analysis of regime shifts in frequency data. This method identifies those points in the development of a process for which deviations are most extreme. Based on a statistical model, functions are estimated that describe the process. This description can represent either the entire series of scores or the series before and after a shift point. The shift point can be either given a priori or estimated from the data. The method is hybrid in that it first uses standard models for the estimation of parameters of the process that is examined and then, in a second step, elements of Configural Frequency Analysis. Uni- and multivariate versions of the method are proposed. In data examples, road traffic data from California and Germany are analyzed before and after particular shift points. Extensions of the proposed method are discussed.