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

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.