Deutsch
 
Datenschutzhinweis Impressum
  DetailsucheBrowse

Datensatz

DATENSATZ AKTIONENEXPORT

Freigegeben

Zeitschriftenartikel

Breakpoint Detection within the Time Series. Modeling Approach Upon Paleoclimatic Proxy Data

Urheber*innen

Lüder,  J.
External Organizations;

/persons/resource/brau

Brauer,  Achim
5.2 Climate Dynamics and Landscape Evolution, 5.0 Earth Surface Processes, Departments, GFZ Publication Database, Deutsches GeoForschungsZentrum;

Jurisch,  R.
External Organizations;

Externe Ressourcen
Es sind keine externen Ressourcen hinterlegt
Volltexte (frei zugänglich)
Es sind keine frei zugänglichen Volltexte in GFZpublic verfügbar
Ergänzendes Material (frei zugänglich)
Es sind keine frei zugänglichen Ergänzenden Materialien verfügbar
Zitation

Lüder, J., Brauer, A., Jurisch, R. (2012): Breakpoint Detection within the Time Series. Modeling Approach Upon Paleoclimatic Proxy Data. - Historical Social Research - Historische Sozialforschung, 37, 2, 315-325.


https://gfzpublic.gfz-potsdam.de/pubman/item/item_245197
Zusammenfassung
»Bruchpunktdetektion im Prozess der Zeitreihenmodellierung anhand paläoklimatologischer Proxydaten«. A large portion of research in time series analysis addresses questioning specific components like trend, cycle or seasonal behavior. Although there is a vast number of publications, only a small amount focuses on the research of irregularities, which are supposed to be within time series – especially in long-term data. Thus this paper focuses on detection of those irregularities and to illustrate the importance of breakpoint estimation. The underlying research theme is given by the discipline of Paleoclimatology. The investigation has been realized upon varved lake sediments as one of common proxydata in paleoclimatics. As the discipline provides insights in climate variability, questioning climate changes implies crucial information about mechanisms of rapid climate shifts. The paper shall also outline the importance of such information, since conducting time series modeling without interdisciplinary research constitutes an almost impossible task. Consequently time series analysis turns out in a procedure of modeling supposed components – like trend, cycles or irregularities – of the underlying datagenerating process and to image those in an appropriate degree.