MULTIVARIATE AR SYSTEMS and MIXED FREQUENCY DATA: G-IDENTIFIABILITY and ESTIMATION

Brian D.O. Anderson; Manfred Deistler; Elisabeth Felsenstein; Bernd Funovits; Lukas Koelbl; Mohsen Zamani

doi:10.1017/S0266466615000043

MULTIVARIATE AR SYSTEMS and MIXED FREQUENCY DATA: G-IDENTIFIABILITY and ESTIMATION

Brian D.O. Anderson, Manfred Deistler^*, Elisabeth Felsenstein, Bernd Funovits, Lukas Koelbl, Mohsen Zamani

^*Corresponding author for this work

Research output: Contribution to journal › Article › peer-review

22 Citations (Scopus)

Abstract

This paper is concerned with the problem of identifiability of the parameters of a high frequency multivariate autoregressive model from mixed frequency time series data. We demonstrate identifiability for generic parameter values using the population second moments of the observations. In addition we display a constructive algorithm for the parameter values and establish the continuity of the mapping attaching the high frequency parameters to these population second moments. These structural results are obtained using two alternative tools viz. extended Yule Walker equations and blocking of the output process. The cases of stock and flow variables, as well as of general linear transformations of high frequency data, are treated. Finally, we briefly discuss how our constructive identifiability results can be used for parameter estimation based on the sample second moments.

Original language	English
Pages (from-to)	793-826
Number of pages	34
Journal	Econometric Theory
Volume	32
Issue number	4
DOIs	https://doi.org/10.1017/S0266466615000043
Publication status	Published - 1 Aug 2016

Access to Document

10.1017/S0266466615000043

Cite this

@article{34980bf25818415e80d3c05df6833fa0,

title = "MULTIVARIATE AR SYSTEMS and MIXED FREQUENCY DATA: G-IDENTIFIABILITY and ESTIMATION",

abstract = "This paper is concerned with the problem of identifiability of the parameters of a high frequency multivariate autoregressive model from mixed frequency time series data. We demonstrate identifiability for generic parameter values using the population second moments of the observations. In addition we display a constructive algorithm for the parameter values and establish the continuity of the mapping attaching the high frequency parameters to these population second moments. These structural results are obtained using two alternative tools viz. extended Yule Walker equations and blocking of the output process. The cases of stock and flow variables, as well as of general linear transformations of high frequency data, are treated. Finally, we briefly discuss how our constructive identifiability results can be used for parameter estimation based on the sample second moments.",

author = "Anderson, {Brian D.O.} and Manfred Deistler and Elisabeth Felsenstein and Bernd Funovits and Lukas Koelbl and Mohsen Zamani",

note = "Publisher Copyright: Copyright {\textcopyright} 2015 Cambridge University Press.",

year = "2016",

month = aug,

day = "1",

doi = "10.1017/S0266466615000043",

language = "English",

volume = "32",

pages = "793--826",

journal = "Econometric Theory",

issn = "0266-4666",

publisher = "Cambridge University Press",

number = "4",

}

TY - JOUR

T1 - MULTIVARIATE AR SYSTEMS and MIXED FREQUENCY DATA

T2 - G-IDENTIFIABILITY and ESTIMATION

AU - Anderson, Brian D.O.

AU - Deistler, Manfred

AU - Felsenstein, Elisabeth

AU - Funovits, Bernd

AU - Koelbl, Lukas

AU - Zamani, Mohsen

PY - 2016/8/1

Y1 - 2016/8/1

N2 - This paper is concerned with the problem of identifiability of the parameters of a high frequency multivariate autoregressive model from mixed frequency time series data. We demonstrate identifiability for generic parameter values using the population second moments of the observations. In addition we display a constructive algorithm for the parameter values and establish the continuity of the mapping attaching the high frequency parameters to these population second moments. These structural results are obtained using two alternative tools viz. extended Yule Walker equations and blocking of the output process. The cases of stock and flow variables, as well as of general linear transformations of high frequency data, are treated. Finally, we briefly discuss how our constructive identifiability results can be used for parameter estimation based on the sample second moments.

AB - This paper is concerned with the problem of identifiability of the parameters of a high frequency multivariate autoregressive model from mixed frequency time series data. We demonstrate identifiability for generic parameter values using the population second moments of the observations. In addition we display a constructive algorithm for the parameter values and establish the continuity of the mapping attaching the high frequency parameters to these population second moments. These structural results are obtained using two alternative tools viz. extended Yule Walker equations and blocking of the output process. The cases of stock and flow variables, as well as of general linear transformations of high frequency data, are treated. Finally, we briefly discuss how our constructive identifiability results can be used for parameter estimation based on the sample second moments.

UR - http://www.scopus.com/inward/record.url?scp=84979734850&partnerID=8YFLogxK

U2 - 10.1017/S0266466615000043

DO - 10.1017/S0266466615000043

M3 - Article

SN - 0266-4666

VL - 32

SP - 793

EP - 826

JO - Econometric Theory

JF - Econometric Theory

IS - 4

ER -

MULTIVARIATE AR SYSTEMS and MIXED FREQUENCY DATA: G-IDENTIFIABILITY and ESTIMATION

Abstract

Access to Document

Other files and links

Fingerprint

Cite this