Time varying VARs with inequality restrictions

Gary Koop, Simon Potter

Research output: Working paper

Abstract

In many applications involving time-varying parameter VARs, it is desirable to restrict the VAR coe¢ cients 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, previous algorithms involve an approximation relating to a key integrating constant. Our algorithms are exact, not involving this approximation. In an application involving a commonly-used U.S. data set, we show how this approximation can be a poor one and present evidence that the algorithms proposed in this paper work well.
LanguageEnglish
Place of PublicationGlasgow
PublisherUniversity of Strathclyde
Publication statusUnpublished - Nov 2008

Keywords

  • bayesian
  • state space model
  • Markov chain montecarlo method
  • time varying parameters

Cite this

Koop, G., & Potter, S. (2008). Time varying VARs with inequality restrictions. Glasgow: University of Strathclyde.
Koop, Gary ; Potter, Simon. / Time varying VARs with inequality restrictions. Glasgow : University of Strathclyde, 2008.
@techreport{9329423058a24c998a44c7bb8cea3c49,
title = "Time varying VARs with inequality restrictions",
abstract = "In many applications involving time-varying parameter VARs, it is desirable to restrict the VAR coe¢ cients 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, previous algorithms involve an approximation relating to a key integrating constant. Our algorithms are exact, not involving this approximation. In an application involving a commonly-used U.S. data set, we show how this approximation can be a poor one and present evidence that the algorithms proposed in this paper work well.",
keywords = "bayesian, state space model, Markov chain montecarlo method, time varying parameters",
author = "Gary Koop and Simon Potter",
year = "2008",
month = "11",
language = "English",
publisher = "University of Strathclyde",
type = "WorkingPaper",
institution = "University of Strathclyde",

}

Koop, G & Potter, S 2008 'Time varying VARs with inequality restrictions' University of Strathclyde, Glasgow.

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

Glasgow : University of Strathclyde, 2008.

Research output: Working paper

TY - UNPB

T1 - Time varying VARs with inequality restrictions

AU - Koop, Gary

AU - Potter, Simon

PY - 2008/11

Y1 - 2008/11

N2 - In many applications involving time-varying parameter VARs, it is desirable to restrict the VAR coe¢ cients 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, previous algorithms involve an approximation relating to a key integrating constant. Our algorithms are exact, not involving this approximation. In an application involving a commonly-used U.S. data set, we show how this approximation can be a poor one and 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 coe¢ cients 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, previous algorithms involve an approximation relating to a key integrating constant. Our algorithms are exact, not involving this approximation. In an application involving a commonly-used U.S. data set, we show how this approximation can be a poor one and present evidence that the algorithms proposed in this paper work well.

KW - bayesian

KW - state space model

KW - Markov chain montecarlo method

KW - time varying parameters

M3 - Working paper

BT - Time varying VARs with inequality restrictions

PB - University of Strathclyde

CY - Glasgow

ER -

Koop G, Potter S. Time varying VARs with inequality restrictions. Glasgow: University of Strathclyde. 2008 Nov.