A data–cleaning augmented Kalman filter for robust estimation of state space models

Marczak, Martyna; Proietti, Tommaso; Grassi, Stefano

Eingang zum Volltext

Marczak, Martyna ; Proietti, Tommaso ; Grassi, Stefano

A data–cleaning augmented Kalman filter for robust estimation of state space models

Bitte beziehen Sie sich beim Zitieren dieses Dokumentes immer auf folgende
URN: urn:nbn:de:bsz:100-opus-11563
URL: http://opus.uni-hohenheim.de/volltexte/2015/1156/

pdf-Format:

Dokument 1.pdf (816 KB)

Gedruckte Ausgabe:

Print-on-Demand-Kopie

Dokument in Google Scholar suchen:

Social Media:

Export:

Abrufstatistik:

SWD-Schlagwörter:

Filterung <Stochastik> , Handelsstatistik

Freie Schlagwörter (Englisch):

robust filtering , augmented Kalman filter , structural time series model , additive outlier , innovation outlier

Institut:

Institut für Volkswirtschaftslehre

DDC-Sachgruppe:

Sozialwissenschaften, Soziologie, Anthropologie

Dokumentart:

ResearchPaper

Schriftenreihe:

Hohenheim discussion papers in business, economics and social sciences

Bandnummer:

2015,13

Sprache:

Englisch

Erstellungsjahr:

2015

Publikationsdatum:

09.11.2015

Lizenz:

Veröffentlichungsvertrag mit der Universitätsbibliothek Hohenheim

Kurzfassung auf Englisch:

This article presents a robust augmented Kalman filter that extends the data– cleaning filter (Masreliez and Martin, 1977) to the general state space model featuring nonstationary and regression effects. The robust filter shrinks the observations towards their one–step–ahead prediction based on the past, by bounding the effect of the information carried by a new observation according to an influence function. When maximum likelihood estimation is carried out on the replacement data, an M–type estimator is obtained. We investigate the performance of the robust AKF in two applications using as a modeling framework the basic structural time series model, a popular unobserved components model in the analysis of seasonal time series. First, a Monte Carlo experiment is conducted in order to evaluate the com- parative accuracy of the proposed method for estimating the variance parameters. Second, the method is applied in a forecasting context to a large set of European trade statistics series.