Structural analysis with Multivariate Autoregressive Index models

Andrea Carriero; George Kapetanios; Massimiliano Marcellino

doi:10.1016/j.jeconom.2016.02.002

Structural analysis with Multivariate Autoregressive Index models

Andrea Carriero, George Kapetanios, Massimiliano Marcellino^*

^*Corresponding author for this work

King's Business School

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

15 Citations (Scopus)

Abstract

We address the issue of parameter dimensionality reduction in Vector Autoregressive models (VARs) for many variables by imposing specific reduced rank restrictions on the coefficient matrices that simplify the VARs into Multivariate Autoregressive Index (MAI) models. We derive the Wold representation implied by the MAIs and show that it is closely related to that associated with dynamic factor models. Then, the theoretical analysis is extended to the case of general rank restrictions on the VAR coefficients. Next, we describe classical and Bayesian estimation of large MAIs, and discuss methods for rank determination. Finally, the performance of the MAIs is compared with that of large Bayesian VARs in the context of Monte Carlo simulations and two empirical applications, on the transmission mechanism of monetary policy and on the propagation of demand and supply shocks.

Original language	English
Pages (from-to)	332-348
Number of pages	17
Journal	JOURNAL OF ECONOMETRICS
Volume	192
Issue number	2
Early online date	11 Feb 2016
DOIs	https://doi.org/10.1016/j.jeconom.2016.02.002
Publication status	Published - 1 Jun 2016

Keywords

Bayesian VARs
C11
C13
C33
C53
Factor models
Forecasting
Large datasets
Multivariate Autoregressive Index models
Reduced rank regressions
Structural analysis

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1016/j.jeconom.2016.02.002

Cite this

@article{2c84bf3c7c0a43e09dbbac9648ef6ce1,

title = "Structural analysis with Multivariate Autoregressive Index models",

abstract = "We address the issue of parameter dimensionality reduction in Vector Autoregressive models (VARs) for many variables by imposing specific reduced rank restrictions on the coefficient matrices that simplify the VARs into Multivariate Autoregressive Index (MAI) models. We derive the Wold representation implied by the MAIs and show that it is closely related to that associated with dynamic factor models. Then, the theoretical analysis is extended to the case of general rank restrictions on the VAR coefficients. Next, we describe classical and Bayesian estimation of large MAIs, and discuss methods for rank determination. Finally, the performance of the MAIs is compared with that of large Bayesian VARs in the context of Monte Carlo simulations and two empirical applications, on the transmission mechanism of monetary policy and on the propagation of demand and supply shocks.",

keywords = "Bayesian VARs, C11, C13, C33, C53, Factor models, Forecasting, Large datasets, Multivariate Autoregressive Index models, Reduced rank regressions, Structural analysis",

author = "Andrea Carriero and George Kapetanios and Massimiliano Marcellino",

year = "2016",

month = jun,

day = "1",

doi = "10.1016/j.jeconom.2016.02.002",

language = "English",

volume = "192",

pages = "332--348",

journal = "JOURNAL OF ECONOMETRICS",

issn = "0304-4076",

publisher = "Elsevier BV",

number = "2",

}

TY - JOUR

T1 - Structural analysis with Multivariate Autoregressive Index models

AU - Carriero, Andrea

AU - Kapetanios, George

AU - Marcellino, Massimiliano

PY - 2016/6/1

Y1 - 2016/6/1

N2 - We address the issue of parameter dimensionality reduction in Vector Autoregressive models (VARs) for many variables by imposing specific reduced rank restrictions on the coefficient matrices that simplify the VARs into Multivariate Autoregressive Index (MAI) models. We derive the Wold representation implied by the MAIs and show that it is closely related to that associated with dynamic factor models. Then, the theoretical analysis is extended to the case of general rank restrictions on the VAR coefficients. Next, we describe classical and Bayesian estimation of large MAIs, and discuss methods for rank determination. Finally, the performance of the MAIs is compared with that of large Bayesian VARs in the context of Monte Carlo simulations and two empirical applications, on the transmission mechanism of monetary policy and on the propagation of demand and supply shocks.

AB - We address the issue of parameter dimensionality reduction in Vector Autoregressive models (VARs) for many variables by imposing specific reduced rank restrictions on the coefficient matrices that simplify the VARs into Multivariate Autoregressive Index (MAI) models. We derive the Wold representation implied by the MAIs and show that it is closely related to that associated with dynamic factor models. Then, the theoretical analysis is extended to the case of general rank restrictions on the VAR coefficients. Next, we describe classical and Bayesian estimation of large MAIs, and discuss methods for rank determination. Finally, the performance of the MAIs is compared with that of large Bayesian VARs in the context of Monte Carlo simulations and two empirical applications, on the transmission mechanism of monetary policy and on the propagation of demand and supply shocks.

KW - Bayesian VARs

KW - C11

KW - C13

KW - C33

KW - C53

KW - Factor models

KW - Forecasting

KW - Large datasets

KW - Multivariate Autoregressive Index models

KW - Reduced rank regressions

KW - Structural analysis

UR - http://www.scopus.com/inward/record.url?scp=84959377796&partnerID=8YFLogxK

U2 - 10.1016/j.jeconom.2016.02.002

DO - 10.1016/j.jeconom.2016.02.002

M3 - Article

AN - SCOPUS:84959377796

SN - 0304-4076

VL - 192

SP - 332

EP - 348

JO - JOURNAL OF ECONOMETRICS

JF - JOURNAL OF ECONOMETRICS

IS - 2

ER -

Structural analysis with Multivariate Autoregressive Index models

Abstract

Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this