Onestep local quasilikelihood estimation

Jianqing Fan Jianwei Chen

Statistics Theory and Methods mathscidoc:1912.43312

Journal of the Royal Statistical Society: Series B (Statistical Methodology), 61, (4), 927-943, 1999
Local quasilikelihood estimation is a useful extension of local least squares methods, but its computational cost and algorithmic convergence problems make the procedure less appealing, particularly when it is iteratively used in methods such as the backfitting algorithm, crossvalidation and bootstrapping. A onestep local quasilikelihood estimator is introduced to overcome the computational drawbacks of the local quasilikelihood method. We demonstrate that as long as the initial estimators are reasonably good, the onestep estimator has the same asymptotic behaviour as the local quasilikelihood method. Our simulation shows that the onestep estimator performs at least as well as the local quasilikelihood method for a wide range of choices of bandwidths. A datadriven bandwidth selector is proposed for the onestep estimator based on the preasymptotic substitution method of Fan and Gijbels. It is
No keywords uploaded!
[ Download ] [ 2019-12-21 11:36:36 uploaded by Jianqing_Fan ] [ 280 downloads ] [ 0 comments ]
  title={Onestep local quasilikelihood estimation},
  author={Jianqing Fan, and Jianwei Chen},
  booktitle={Journal of the Royal Statistical Society: Series B (Statistical Methodology)},
Jianqing Fan, and Jianwei Chen. Onestep local quasilikelihood estimation. 1999. Vol. 61. In Journal of the Royal Statistical Society: Series B (Statistical Methodology). pp.927-943.
Please log in for comment!
Contact us: | Copyright Reserved