Recursive Residuals and Model Diagnostics for Normal and Non-Normal State Space Models

Sylvia Frühwirth-Schnatter

doi:10.57938/469a9e44-ed5e-4e96-9004-206f0b9545fd

Recursive Residuals and Model Diagnostics for Normal and Non-Normal State Space Models

Sylvia Frühwirth-Schnatter

Institute for Statistics and Mathematics

Publication: Working/Discussion Paper › WU Working Paper

38 Downloads (Pure)

Abstract

Model diagnostics for normal and non-normal state space models is based on recursive residuals which are defined from the one-step ahead predictive distribution. Routine calculation of these residuals is discussed in detail. Various tools of diagnostics are suggested to check e.g. for wrong observation distributions and for autocorrelation. The paper also covers such topics as model diagnostics for discrete time series, model diagnostics for generalized linear models, and model discrimination via Bayes factors.

Original language	English
Place of Publication	Vienna
Publisher	Department of Statistics and Mathematics, WU Vienna University of Economics and Business
DOIs	https://doi.org/10.57938/469a9e44-ed5e-4e96-9004-206f0b9545fd
Publication status	Published - 1994

Publication series

Series	Forschungsberichte / Institut für Statistik
Number	40

WU Working Paper Series

Forschungsberichte / Institut für Statistik

Access to Document

10.57938/469a9e44-ed5e-4e96-9004-206f0b9545fdLicence: Unspecified

DocumentFinal published version, 1.2 MBLicence: Unspecified

Check availability

University Library Catalogue

1 Journal article

Recursive residuals and model diagnostics for normal and nonnormal state space models
Frühwirth-Schnatter, S., 1996, In: Environmental and Ecological Statistics. 3, p. 291 - 309
Publication: Scientific journal › Journal article › peer-review

Cite this

@techreport{469a9e44ed5e4e969004206f0b9545fd,

title = "Recursive Residuals and Model Diagnostics for Normal and Non-Normal State Space Models",

abstract = "Model diagnostics for normal and non-normal state space models is based on recursive residuals which are defined from the one-step ahead predictive distribution. Routine calculation of these residuals is discussed in detail. Various tools of diagnostics are suggested to check e.g. for wrong observation distributions and for autocorrelation. The paper also covers such topics as model diagnostics for discrete time series, model diagnostics for generalized linear models, and model discrimination via Bayes factors.",

author = "Sylvia Fr{\"u}hwirth-Schnatter",

year = "1994",

doi = "10.57938/469a9e44-ed5e-4e96-9004-206f0b9545fd",

language = "English",

series = "Forschungsberichte / Institut f{\"u}r Statistik",

number = "40",

publisher = "Department of Statistics and Mathematics, WU Vienna University of Economics and Business",

edition = "February 1994",

type = "WorkingPaper",

institution = "Department of Statistics and Mathematics, WU Vienna University of Economics and Business",

}

TY - UNPB

T1 - Recursive Residuals and Model Diagnostics for Normal and Non-Normal State Space Models

AU - Frühwirth-Schnatter, Sylvia

PY - 1994

Y1 - 1994

N2 - Model diagnostics for normal and non-normal state space models is based on recursive residuals which are defined from the one-step ahead predictive distribution. Routine calculation of these residuals is discussed in detail. Various tools of diagnostics are suggested to check e.g. for wrong observation distributions and for autocorrelation. The paper also covers such topics as model diagnostics for discrete time series, model diagnostics for generalized linear models, and model discrimination via Bayes factors.

AB - Model diagnostics for normal and non-normal state space models is based on recursive residuals which are defined from the one-step ahead predictive distribution. Routine calculation of these residuals is discussed in detail. Various tools of diagnostics are suggested to check e.g. for wrong observation distributions and for autocorrelation. The paper also covers such topics as model diagnostics for discrete time series, model diagnostics for generalized linear models, and model discrimination via Bayes factors.

U2 - 10.57938/469a9e44-ed5e-4e96-9004-206f0b9545fd

DO - 10.57938/469a9e44-ed5e-4e96-9004-206f0b9545fd

M3 - WU Working Paper

T3 - Forschungsberichte / Institut für Statistik

BT - Recursive Residuals and Model Diagnostics for Normal and Non-Normal State Space Models

PB - Department of Statistics and Mathematics, WU Vienna University of Economics and Business

CY - Vienna

ER -

Recursive Residuals and Model Diagnostics for Normal and Non-Normal State Space Models

Abstract

Publication series

WU Working Paper Series

Access to Document

Check availability

Other versions

Recursive residuals and model diagnostics for normal and nonnormal state space models

Cite this