Bitte verwenden Sie diesen Link, um diese Publikation zu zitieren, oder auf sie als Internetquelle zu verweisen: https://hdl.handle.net/10419/195462 
Erscheinungsjahr: 
2018
Quellenangabe: 
[Journal:] Econometrics [ISSN:] 2225-1146 [Volume:] 6 [Issue:] 3 [Publisher:] MDPI [Place:] Basel [Year:] 2018 [Pages:] 1-33
Verlag: 
MDPI, Basel
Zusammenfassung: 
Econometric analysis requires filtering techniques that are adapted to cater to data sequences that are short and that have strong trends. Whereas the economists have tended to conduct their analyses in the time domain, the engineers have emphasised the frequency domain. This paper places its emphasis in the frequency domain; and it shows how the frequency-domain methods can be adapted to cater to short trended sequences. Working in the frequency domain allows an unrestricted choice to be made of the frequency response of a filter. It also requires that the data should be free of trends. Methods for extracting the trends prior to filtering and for restoring them thereafter are described.
Schlagwörter: 
time series
Fourier analysis
sampling
filtering
JEL: 
C22
Persistent Identifier der Erstveröffentlichung: 
Creative-Commons-Lizenz: 
cc-by Logo
Dokumentart: 
Article

Datei(en):
Datei
Größe
430.78 kB





Publikationen in EconStor sind urheberrechtlich geschützt.