Time varying VARs with inequality restrictions

Gary Koop, Simon M. Potter

Research output: Contribution to journalArticle

22 Citations (Scopus)

Abstract

In many applications involving time-varying parameter VARs, it is desirable to restrict the VAR coefficients at each point in time to be non-explosive. This is an example of a problem where inequality restrictions are imposed on states in a state space model. In this paper, we describe how existing MCMC algorithms for imposing such inequality restrictions can work poorly (or not at all) and suggest alternative algorithms which exhibit better performance. Furthermore, we show that previous algorithms involve an approximation relating to a key prior integrating constant. Our algorithms are exact, not involving this approximation. In an application involving a commonly used U.S. data set, we present evidence that the algorithms proposed in this paper work well.
LanguageEnglish
Pages1126-1138
Number of pages13
JournalJournal of Economic Dynamics and Control
Volume35
Issue number7
DOIs
Publication statusPublished - Jul 2011

Fingerprint

Time-varying
Restriction
MCMC Algorithm
Time-varying Parameters
State-space Model
Approximation
Inequality restrictions
Alternatives
Coefficient

Keywords

  • Bayesian
  • state space model
  • Markov chain Monte Carlo
  • Metropolis–Hastings

Cite this

Koop, Gary ; Potter, Simon M. / Time varying VARs with inequality restrictions. In: Journal of Economic Dynamics and Control. 2011 ; Vol. 35, No. 7. pp. 1126-1138.
@article{64fcd6be3ba444c49052ed2597a64d08,
title = "Time varying VARs with inequality restrictions",
abstract = "In many applications involving time-varying parameter VARs, it is desirable to restrict the VAR coefficients at each point in time to be non-explosive. This is an example of a problem where inequality restrictions are imposed on states in a state space model. In this paper, we describe how existing MCMC algorithms for imposing such inequality restrictions can work poorly (or not at all) and suggest alternative algorithms which exhibit better performance. Furthermore, we show that previous algorithms involve an approximation relating to a key prior integrating constant. Our algorithms are exact, not involving this approximation. In an application involving a commonly used U.S. data set, we present evidence that the algorithms proposed in this paper work well.",
keywords = "Bayesian, state space model, Markov chain Monte Carlo, Metropolis–Hastings",
author = "Gary Koop and Potter, {Simon M.}",
year = "2011",
month = "7",
doi = "10.1016/j.jedc.2011.02.001",
language = "English",
volume = "35",
pages = "1126--1138",
journal = "Journal of Economic Dynamics and Control",
issn = "0165-1889",
number = "7",

}

Time varying VARs with inequality restrictions. / Koop, Gary; Potter, Simon M.

In: Journal of Economic Dynamics and Control, Vol. 35, No. 7, 07.2011, p. 1126-1138.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Time varying VARs with inequality restrictions

AU - Koop, Gary

AU - Potter, Simon M.

PY - 2011/7

Y1 - 2011/7

N2 - In many applications involving time-varying parameter VARs, it is desirable to restrict the VAR coefficients at each point in time to be non-explosive. This is an example of a problem where inequality restrictions are imposed on states in a state space model. In this paper, we describe how existing MCMC algorithms for imposing such inequality restrictions can work poorly (or not at all) and suggest alternative algorithms which exhibit better performance. Furthermore, we show that previous algorithms involve an approximation relating to a key prior integrating constant. Our algorithms are exact, not involving this approximation. In an application involving a commonly used U.S. data set, we present evidence that the algorithms proposed in this paper work well.

AB - In many applications involving time-varying parameter VARs, it is desirable to restrict the VAR coefficients at each point in time to be non-explosive. This is an example of a problem where inequality restrictions are imposed on states in a state space model. In this paper, we describe how existing MCMC algorithms for imposing such inequality restrictions can work poorly (or not at all) and suggest alternative algorithms which exhibit better performance. Furthermore, we show that previous algorithms involve an approximation relating to a key prior integrating constant. Our algorithms are exact, not involving this approximation. In an application involving a commonly used U.S. data set, we present evidence that the algorithms proposed in this paper work well.

KW - Bayesian

KW - state space model

KW - Markov chain Monte Carlo

KW - Metropolis–Hastings

UR - http://www.scopus.com/inward/record.url?scp=79955966149&partnerID=8YFLogxK

U2 - 10.1016/j.jedc.2011.02.001

DO - 10.1016/j.jedc.2011.02.001

M3 - Article

VL - 35

SP - 1126

EP - 1138

JO - Journal of Economic Dynamics and Control

T2 - Journal of Economic Dynamics and Control

JF - Journal of Economic Dynamics and Control

SN - 0165-1889

IS - 7

ER -