On the identifiability of minimal VARMA representations

DOI10.1023/A:1009955223247MaRDI QIDQ1281920zbMATH OpenOpenAlexFDO

Authors C. Francq, Alain Berlinet

Publication date 16 August 1999

Published in Statistical Inference for Stochastic Processes (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1023/a:1009955223247

zbMATH Keywords

identifiabilityidentificationminimal ordersmultivariate ARMA models

Mathematics Subject Classification ID

Time series, auto-correlation, regression, etc. in statistics (GARCH) (62M10)

Review text

It is well known that, in general, the minimal orders of VARMA processes are not unique. For example, a VARMA process may admit both an AR(1) and an MA(1) representation without being a white noise. The authors propose a way of finding all the minimal orders of the representations of VARMA processes in terms of the autocovariance function and a way of checking whether a representation with minimal orders is identifiable or not. Examples of VARMA processes with zero, one, or several standard identifiable representations are introduced.

Recommendations

• scientific article; zbMATH DE number 4111699
• scientific article; zbMATH DE number 3996943
• scientific article; zbMATH DE number 1031926
• VARMA representation of DSGE models
• Identification of multivariate ARMA models

Cited in

(5)

• VARMA representation of DSGE models
• scientific article; zbMATH DE number 1031926 (Why is no real title available?)
• scientific article; zbMATH DE number 3996943 (Why is no real title available?)
• Identifiability conditions for generalised STARMA models
• Identifiability and minimality in rational models

This page was built for publication: On the identifiability of minimal VARMA representations

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1281920)