A note on moment convergence of bootstrap M-estimators
A note on moment convergence of bootstrap M-estimators
复制标题
DOI：
10.1524/stnd.2011.1078
复制
发表时间：
2011-03-01
影响因子：
1.5
通讯作者：
Kato, Kengo
中科院分区：
文献类型：
--
作者：
Kato, Kengo
本文研究了一般M-估计的Bootstrap矩估计的相合性。建立了Bootstrap M-估计的一致可积性定理，从而给出了Bootstrap矩估计相合的充分条件。作为我们定理的应用，我们为分位数回归估计量的自助方差估计量的相容性提供了充分条件，这一直被认为是文献中未解决的重要问题。我们还讨论了自举信息准则的理由。
This paper studies the consistency of bootstrap moment estimators for a general M-estimator. We establish a theorem on the uniform integrability of the bootstrap M-estimator, thereby giving sufficient conditions for the consistency of the bootstrap moment estimators. As an application of our theorem, we provide sufficient conditions for the consistency of the bootstrap variance estimator for the quantile regression estimator, which has been considered as an important unsolved problem in the literature. We also discuss a justification of a bootstrap information criterion.