Initial conditions and moment restrictions in dynamic panel data models

Abstract

Estimation of the dynamic error components model is considered using two alternative linear estimators that are designed to improve the properties of the standard firstdifferenced GMM estimator. Both estimators require restrictions on the initial conditions process. Asymptotic efficiency comparisons and Monte Carlo simulations for the simple AR(1) model demonstrate the dramatic improvement in performance of the proposed estimators compared to the usual first-differenced GMM estimator, and compared to non-linear GMM. The importance of these results is illustrated in an application to the estimation of a labour demand model using company panel data.1998 Elsevier Science S.A. All rights reserved. JEL classification: C23

Foreign Direct Investment, Macroeconomic Instability and Economic Growth in Mena Countries

“…We assessed the growth-effect of FDI, using data from MENA countries on period 1970-2005. To deal properly with dynamic panel models, we use GMM estimators designed by Arellano and Bond (1991), Blundell and Bond (1998), and 2SLS estimators designed by Anderson and Hsiao (1982).…”

“…It also accounts for the bias induced by including the lagged real per capita GDP in the equation of growth 6 . Our strategy for estimation uses the Generalized Method of Moments (GMM) estimators suggested for the dynamics of adjustment that were developed by Arellano and Bond (1991), and Blundell and Bond (1998). To analyse the sensibility of our results to the GMM method, we also use the standard Two-stage Least Squares (2SLS) estimators developed by Anderson and Hsiao (1982).…”

“…But , in order to work in the GMM framework, using deeper lags of as additional instruments, we use both classic Arellano-Bond (1991) difference andBlundell-Bond (1998) system estimators for dynamic panels. These estimators use a larger set of moment conditions.…”

“…The authors are sceptical because these previous studies did not fully control for simultaneity bias, country-specific effects, and the use of routine of lagged dependant variable in growth regressions 2 . In order to estimate consistent and efficient parameters, Carkovic and Levine used the Generalized Method of Moments (GMM) panel estimators 3 designed by Arellano and Bond (1991), Arellano and Bover (1995), and Blundell and Bond (1998). Our paper integrates to certain extent some methodological elements used in the study of Carkovic and Levine.…”

Economic Growth Volatility and Resource Curse: The Role of Financial Development

“…Blundell and Bond (1998) instead, suggested using the system GMM estimator developed by Arellano and Bover (1995) to deal with weak instrumental variables. They included the different forms of the lagged dependent variable into the matrix of instrumental variables.…”