Effcient M-estimators with auxiliary information

This paper introduces a new class of M-estimators based on generalised empirical likelihood estimation with some auxiliary information available in the sample. The resulting class of estimators is efficient in the sense that it achieves the same asymptotic lower bound as that of the efficient generalised method of moment-based M-estimator with the same auxiliary information. The results of the paper are quite general and apply to M-estimators defined by both smooth and nonsmooth estimating equations. Simulations show that the proposed estimators perform well in finite samples, and can be less biased and more precise than standard M-estimators within China.