A flexible approach to parametric inference in nonlinear time series models

G.M. Koop, S. Potter

Research output: Working paper

Abstract

Many structural break and regime-switching models have been used with macroeconomic and …nancial data. In this paper, we develop an extremely flexible parametric model which can accommodate virtually any of these speci…cations and does so in a simple way which allows for straightforward Bayesian inference. The basic idea underlying our model is that it adds two simple concepts to a standard state space framework. These ideas are ordering and distance. By ordering the data in various ways, we can accommodate a wide variety of nonlinear time series models, including those with regime-switching and structural breaks. By allowing the state equation variances to depend on the distance between observations, the parameters can evolve in a wide variety of ways, allowing for everything from models exhibiting abrupt change (e.g. threshold autoregressive models or standard structural break models) to those which allow for a gradual evolution of parameters (e.g. smooth transition autoregressive models or time varying parameter models). We show how our model will (approximately) nest virtually every popular model in the regime-switching and structural break literatures. Bayesian econometric methods for inference in this model are developed. Because we stay within a state space framework, these methods are relatively straightforward, drawing on the existing literature. We use arti…cial data to show the advantages of our approach, before providing two empirical illustrations involving the modeling of real GDP growth.
LanguageEnglish
Place of PublicationGlasgow
PublisherUniversity of Strathclyde
Number of pages38
Publication statusUnpublished - Dec 2005

Fingerprint

Nonlinear time series
Inference
Time series models
Structural breaks
Regime switching
State space
Threshold autoregressive model
Smooth transition
GDP growth
Parametric model
Modeling
Macroeconomics
Time-varying parameter model
Bayesian inference
Regime-switching model
Bayesian econometrics
Autoregressive model
Econometric methods
Real GDP

Keywords

  • bayesian
  • structural break
  • threshold autoregressive
  • regime switching
  • state space model economics

Cite this

Koop, G. M., & Potter, S. (2005). A flexible approach to parametric inference in nonlinear time series models. Glasgow: University of Strathclyde.
Koop, G.M. ; Potter, S. / A flexible approach to parametric inference in nonlinear time series models. Glasgow : University of Strathclyde, 2005.
@techreport{5c0efff0233c4cc18edae1405bd12f9e,
title = "A flexible approach to parametric inference in nonlinear time series models",
abstract = "Many structural break and regime-switching models have been used with macroeconomic and …nancial data. In this paper, we develop an extremely flexible parametric model which can accommodate virtually any of these speci…cations and does so in a simple way which allows for straightforward Bayesian inference. The basic idea underlying our model is that it adds two simple concepts to a standard state space framework. These ideas are ordering and distance. By ordering the data in various ways, we can accommodate a wide variety of nonlinear time series models, including those with regime-switching and structural breaks. By allowing the state equation variances to depend on the distance between observations, the parameters can evolve in a wide variety of ways, allowing for everything from models exhibiting abrupt change (e.g. threshold autoregressive models or standard structural break models) to those which allow for a gradual evolution of parameters (e.g. smooth transition autoregressive models or time varying parameter models). We show how our model will (approximately) nest virtually every popular model in the regime-switching and structural break literatures. Bayesian econometric methods for inference in this model are developed. Because we stay within a state space framework, these methods are relatively straightforward, drawing on the existing literature. We use arti…cial data to show the advantages of our approach, before providing two empirical illustrations involving the modeling of real GDP growth.",
keywords = "bayesian, structural break, threshold autoregressive, regime switching, state space model economics",
author = "G.M. Koop and S. Potter",
year = "2005",
month = "12",
language = "English",
publisher = "University of Strathclyde",
type = "WorkingPaper",
institution = "University of Strathclyde",

}

Koop, GM & Potter, S 2005 'A flexible approach to parametric inference in nonlinear time series models' University of Strathclyde, Glasgow.

A flexible approach to parametric inference in nonlinear time series models. / Koop, G.M.; Potter, S.

Glasgow : University of Strathclyde, 2005.

Research output: Working paper

TY - UNPB

T1 - A flexible approach to parametric inference in nonlinear time series models

AU - Koop, G.M.

AU - Potter, S.

PY - 2005/12

Y1 - 2005/12

N2 - Many structural break and regime-switching models have been used with macroeconomic and …nancial data. In this paper, we develop an extremely flexible parametric model which can accommodate virtually any of these speci…cations and does so in a simple way which allows for straightforward Bayesian inference. The basic idea underlying our model is that it adds two simple concepts to a standard state space framework. These ideas are ordering and distance. By ordering the data in various ways, we can accommodate a wide variety of nonlinear time series models, including those with regime-switching and structural breaks. By allowing the state equation variances to depend on the distance between observations, the parameters can evolve in a wide variety of ways, allowing for everything from models exhibiting abrupt change (e.g. threshold autoregressive models or standard structural break models) to those which allow for a gradual evolution of parameters (e.g. smooth transition autoregressive models or time varying parameter models). We show how our model will (approximately) nest virtually every popular model in the regime-switching and structural break literatures. Bayesian econometric methods for inference in this model are developed. Because we stay within a state space framework, these methods are relatively straightforward, drawing on the existing literature. We use arti…cial data to show the advantages of our approach, before providing two empirical illustrations involving the modeling of real GDP growth.

AB - Many structural break and regime-switching models have been used with macroeconomic and …nancial data. In this paper, we develop an extremely flexible parametric model which can accommodate virtually any of these speci…cations and does so in a simple way which allows for straightforward Bayesian inference. The basic idea underlying our model is that it adds two simple concepts to a standard state space framework. These ideas are ordering and distance. By ordering the data in various ways, we can accommodate a wide variety of nonlinear time series models, including those with regime-switching and structural breaks. By allowing the state equation variances to depend on the distance between observations, the parameters can evolve in a wide variety of ways, allowing for everything from models exhibiting abrupt change (e.g. threshold autoregressive models or standard structural break models) to those which allow for a gradual evolution of parameters (e.g. smooth transition autoregressive models or time varying parameter models). We show how our model will (approximately) nest virtually every popular model in the regime-switching and structural break literatures. Bayesian econometric methods for inference in this model are developed. Because we stay within a state space framework, these methods are relatively straightforward, drawing on the existing literature. We use arti…cial data to show the advantages of our approach, before providing two empirical illustrations involving the modeling of real GDP growth.

KW - bayesian

KW - structural break

KW - threshold autoregressive

KW - regime switching

KW - state space model economics

M3 - Working paper

BT - A flexible approach to parametric inference in nonlinear time series models

PB - University of Strathclyde

CY - Glasgow

ER -

Koop GM, Potter S. A flexible approach to parametric inference in nonlinear time series models. Glasgow: University of Strathclyde. 2005 Dec.