Further Results on the Weak Instruments Problem of the System GMM Estimator in Dynamic Panel Data Models
研究了动态面板数据模型中系统GMM估计量的弱工具变量问题，提出前向GLS变换可提高集中参数，从而增强工具变量强度，蒙特卡洛模拟显示新估计量优于传统系统GMM估计量。
Abstract In this paper, we investigate the weak instruments problem of the generalized method of moments (GMM) estimator for dynamic panel data models. Specifically, we complement Bun and Windmeijer (2010) by considering the alternative first‐difference and level models transformed by the forward GLS transformation. We demonstrate that this transformation yields a higher concentration parameter compared with the original models. This indicates that the proposed transformation yields stronger instruments even though the instruments used are identical. The Monte Carlo simulation results show that the system GMM estimator for the transformed model, called the forward system GMM estimator , performs better than the conventional system GMM estimator.