摘要：The Arellano and Bond (1991) estimator is widely-used among applied researchers whenestimating dynamic panels with fi xed effects and predetermined regressors. This estimatormight behave poorly in fi nite samples when the cross-section dimension of the data is small(i.e. small N), especially if the variables under analysis are persistent over time. This paperdiscusses a maximum likelihood estimator that is asymptotically equivalent to Arellano andBond (1991) but presents better fi nite sample behaviour. Moreover, the estimator is easyto implement in Stata using the xtdpdml command as described in the companion paperWilliams et al. (2016), which also discusses further advantages of the proposed estimatorfor practitioners.
关键词：dynamic panel data; maximum likelihood estimation.
