数学物理学报(英文版) ›› 1998, Vol. 18 ›› Issue (S1): 68-77.

• 论文 • 上一篇    下一篇

SEMIPARAMETRIC REGRESSION MODELS WITH LOCALLY GENERALIZED GAUSSIAN ERROR'S STRUCTURE

胡舒合   

  1. Department of Mathematics, anhui University, Hefei 230039, China
  • 收稿日期:1997-03-17 修回日期:1997-10-05 出版日期:1998-12-31 发布日期:1998-12-31
  • 基金资助:
    This work is supported by the National Natural Science Foundation of China

SEMIPARAMETRIC REGRESSION MODELS WITH LOCALLY GENERALIZED GAUSSIAN ERROR'S STRUCTURE

Hu Shuhe   

  1. Department of Mathematics, anhui University, Hefei 230039, China
  • Received:1997-03-17 Revised:1997-10-05 Online:1998-12-31 Published:1998-12-31
  • Supported by:
    This work is supported by the National Natural Science Foundation of China

摘要: This paper proposes parametric component and nonparametric component estimators in a semiparametric regression models based on least squares and weight function's method, their strong consistency and rib mean consistency are obtained under a locally generallied Gaussinan error's structure. Finally, the author showes that the usual weight functions based on nearest neighbor method satisfy the deigned assumptions imposed.

关键词: Semiparametric regression, Locally generalized Garussian error, Strong consistency, Rib mean consistency

Abstract: This paper proposes parametric component and nonparametric component estimators in a semiparametric regression models based on least squares and weight function's method, their strong consistency and rib mean consistency are obtained under a locally generallied Gaussinan error's structure. Finally, the author showes that the usual weight functions based on nearest neighbor method satisfy the deigned assumptions imposed.

Key words: Semiparametric regression, Locally generalized Garussian error, Strong consistency, Rib mean consistency