Accuracy and Efficiency of Various GMM Inference Techniques in Dynamic Micro Panel Data Models

58 Pages Posted: 5 Feb 2015

See all articles by Jan F. Kiviet

Jan F. Kiviet

University of Amsterdam - Department of Quantitative Economics; Tinbergen Institute; University of Amsterdam, Dept. of Quantitative Economics

Milan Pleus

University of Amsterdam

Rutger Poldermans

University of Amsterdam

Date Written: January 31, 2015

Abstract

The performance in finite samples is examined of inference obtained by variants of the Arellano-Bond and the Blundell-Bond GMM estimation techniques for single dynamic panel data models with possibly endogenous regressors and cross-sectional heteroskedasticity. By simulation the effects are examined of using particular instrument strength enhancing reductions and transformations of the matrix of instrumental variables, of less robust implementations of the GMM weighting matrix, and also of corrections to the standard asymptotic variance estimates. We compare the root mean squared errors of the coefficient estimators and also the size of tests on coefficient values and of different implementations of overidentification restriction tests. Also the size and power of tests on the validity of the additional orthogonality conditions exploited by the Blundell-Bond technique are assessed over a pretty wide grid of relevant cases. Surprisingly, particular asymptotically optimal and relatively robust weighting matrices are found to be superior in finite samples to ostensibly more appropriate versions. Most of the variants of tests for overidentification restrictions show serious deficiencies. A recently developed modification of GMM is found to have great potential when the cross-sectional heteroskedasticity is pronounced and the time-series dimension of the sample not too small. Finally all techniques are employed to actual data and lead to some profound insights.

Keywords: cross-sectional heteroskedasticity, Sargan-Hansen (incremental) tests, variants of t-tests, weighting matrices, Windmeijer-correction

JEL Classification: C120, C130, C150, C230, C260, C520

Suggested Citation

Kiviet, Jan F. and Pleus, Milan and Poldermans, Rutger, Accuracy and Efficiency of Various GMM Inference Techniques in Dynamic Micro Panel Data Models (January 31, 2015). CESifo Working Paper Series No. 5189, Available at SSRN: https://ssrn.com/abstract=2560749 or http://dx.doi.org/10.2139/ssrn.2560749

Jan F. Kiviet (Contact Author)

University of Amsterdam - Department of Quantitative Economics ( email )

Valckenierstraat 65-67
Amsterdam, 1018 XE
Netherlands
+31(20)5254224 (Phone)
+31(20)5254349 (Fax)

Tinbergen Institute ( email )

Gustav Mahlerplein 117
Amsterdam, 1082 MS
Netherlands

University of Amsterdam, Dept. of Quantitative Economics ( email )

Valckenierstraat 65-67
Amsterdam, 1018 XE
Netherlands

HOME PAGE: http://www.feb.uva.nl/ke/jfk.htm

Milan Pleus

University of Amsterdam ( email )

Spui 21
Amsterdam, 1018 WB
Netherlands

Rutger Poldermans

University of Amsterdam ( email )

Spui 21
Amsterdam, 1018 WB
Netherlands

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
151
Abstract Views
820
Rank
350,945
PlumX Metrics