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

Marczak, Martyna; Proietti, Tommaso; Grassi, Stefano

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

dc.contributor.author	Marczak, Martyna	de
dc.contributor.author	Proietti, Tommaso	de
dc.contributor.author	Grassi, Stefano	de
dc.date.accessioned	2024-04-08T08:51:59Z
dc.date.available	2024-04-08T08:51:59Z
dc.date.created	2015-11-09
dc.date.issued	2015
dc.description.abstract	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.	en
dc.identifier.swb	451518292
dc.identifier.uri	https://hohpublica.uni-hohenheim.de/handle/123456789/5965
dc.identifier.urn	urn:nbn:de:bsz:100-opus-11563
dc.language.iso	eng
dc.relation.ispartofseries	Hohenheim discussion papers in business, economics and social sciences; 2015,13
dc.rights.license	publ-mit-pod	en
dc.rights.license	publ-mit-pod	de
dc.rights.uri	http://opus.uni-hohenheim.de/doku/lic_mit_pod.php
dc.subject	Robust filtering	en
dc.subject	Augmented Kalman filter	en
dc.subject	Structural time series model	en
dc.subject	Additive outlier	en
dc.subject	Innovation outlier	en
dc.subject.ddc	300
dc.subject.gnd	Filterung <Stochastik>	de
dc.subject.gnd	Handelsstatistik	de
dc.title	A data–cleaning augmented Kalman filter for robust estimation of state space models	de
dc.type.dcmi	Text	de
dc.type.dini	WorkingPaper	de
local.access	uneingeschränkter Zugriff	en
local.access	uneingeschränkter Zugriff	de
local.bibliographicCitation.publisherPlace	Universität Hohenheim	de
local.faculty.number	3	de
local.institute.number	520	de
local.opus.number	1156
local.series.issueNumber	2015,13
local.series.title	Hohenheim discussion papers in business, economics and social sciences
local.university	Universität Hohenheim	de
local.university.faculty	Faculty of Business, Economics and Social Sciences	en
local.university.faculty	Fakultät Wirtschafts- und Sozialwissenschaften	de
local.university.institute	Institute for Economics	en
local.university.institute	Institut für Volkswirtschaftslehre	de

Files

Original bundle

Now showing 1 - 1 of 1

Name:: dp_13_2015_online.pdf
Size:: 815.72 KB
Format:: Adobe Portable Document Format
Description:: Open Access Fulltext

Download

Collections

Institut für Volkswirtschaftslehre