A Comparison of Two Averaging Techniques with an Application to Growth Empirics

J.R. Magnus, O.R. Powell, P. Prüfer

Research output: Working paperDiscussion paperOther research output

270 Downloads (Pure)

Abstract

Empirical growth research faces a high degree of model uncertainty. Apart from the neoclassical growth model, many new (endogenous) growth models have been proposed. This causes a lack of robustness of the parameter estimates and makes the determination of the key determinants of growth hazardous. The current paper deals with the fundamental issue of parameter estimation under model uncertainty, and compares the performance of various model averaging techniques. In particular, it contrasts Bayesian model averaging (BMA) — currently one of the standard methods used in growth empirics — with weighted-average least squares (WALS), a method that has not previously been applied in this context.
Original languageEnglish
Place of PublicationTilburg
PublisherEconometrics
Number of pages42
Volume2008-39
Publication statusPublished - 2008

Publication series

NameCentER Discussion Paper
Volume2008-39

Fingerprint

Empirics
Model uncertainty
Endogenous growth model
Bayesian model averaging
Neoclassical growth model
Model averaging
Least squares
Robustness
Parameter estimation

Keywords

  • Model averaging
  • Bayesian analysis
  • Growth determinants

Cite this

Magnus, J. R., Powell, O. R., & Prüfer, P. (2008). A Comparison of Two Averaging Techniques with an Application to Growth Empirics. (CentER Discussion Paper; Vol. 2008-39). Tilburg: Econometrics.
Magnus, J.R. ; Powell, O.R. ; Prüfer, P. / A Comparison of Two Averaging Techniques with an Application to Growth Empirics. Tilburg : Econometrics, 2008. (CentER Discussion Paper).
@techreport{0392dffa51e04bc99644fe8ceff8538e,
title = "A Comparison of Two Averaging Techniques with an Application to Growth Empirics",
abstract = "Empirical growth research faces a high degree of model uncertainty. Apart from the neoclassical growth model, many new (endogenous) growth models have been proposed. This causes a lack of robustness of the parameter estimates and makes the determination of the key determinants of growth hazardous. The current paper deals with the fundamental issue of parameter estimation under model uncertainty, and compares the performance of various model averaging techniques. In particular, it contrasts Bayesian model averaging (BMA) — currently one of the standard methods used in growth empirics — with weighted-average least squares (WALS), a method that has not previously been applied in this context.",
keywords = "Model averaging, Bayesian analysis, Growth determinants",
author = "J.R. Magnus and O.R. Powell and P. Pr{\"u}fer",
note = "Subsequently published in Journal of Econometrics, 2010 Pagination: 42",
year = "2008",
language = "English",
volume = "2008-39",
series = "CentER Discussion Paper",
publisher = "Econometrics",
type = "WorkingPaper",
institution = "Econometrics",

}

Magnus, JR, Powell, OR & Prüfer, P 2008 'A Comparison of Two Averaging Techniques with an Application to Growth Empirics' CentER Discussion Paper, vol. 2008-39, Econometrics, Tilburg.

A Comparison of Two Averaging Techniques with an Application to Growth Empirics. / Magnus, J.R.; Powell, O.R.; Prüfer, P.

Tilburg : Econometrics, 2008. (CentER Discussion Paper; Vol. 2008-39).

Research output: Working paperDiscussion paperOther research output

TY - UNPB

T1 - A Comparison of Two Averaging Techniques with an Application to Growth Empirics

AU - Magnus, J.R.

AU - Powell, O.R.

AU - Prüfer, P.

N1 - Subsequently published in Journal of Econometrics, 2010 Pagination: 42

PY - 2008

Y1 - 2008

N2 - Empirical growth research faces a high degree of model uncertainty. Apart from the neoclassical growth model, many new (endogenous) growth models have been proposed. This causes a lack of robustness of the parameter estimates and makes the determination of the key determinants of growth hazardous. The current paper deals with the fundamental issue of parameter estimation under model uncertainty, and compares the performance of various model averaging techniques. In particular, it contrasts Bayesian model averaging (BMA) — currently one of the standard methods used in growth empirics — with weighted-average least squares (WALS), a method that has not previously been applied in this context.

AB - Empirical growth research faces a high degree of model uncertainty. Apart from the neoclassical growth model, many new (endogenous) growth models have been proposed. This causes a lack of robustness of the parameter estimates and makes the determination of the key determinants of growth hazardous. The current paper deals with the fundamental issue of parameter estimation under model uncertainty, and compares the performance of various model averaging techniques. In particular, it contrasts Bayesian model averaging (BMA) — currently one of the standard methods used in growth empirics — with weighted-average least squares (WALS), a method that has not previously been applied in this context.

KW - Model averaging

KW - Bayesian analysis

KW - Growth determinants

M3 - Discussion paper

VL - 2008-39

T3 - CentER Discussion Paper

BT - A Comparison of Two Averaging Techniques with an Application to Growth Empirics

PB - Econometrics

CY - Tilburg

ER -

Magnus JR, Powell OR, Prüfer P. A Comparison of Two Averaging Techniques with an Application to Growth Empirics. Tilburg: Econometrics. 2008. (CentER Discussion Paper).