摘要
对非参数回归模型yi=g(xi)+ei(i=1,2,…,n),具体讨论误差为NA序列时,对g(x)给出一种加权核估计gn(x),研究gn(x)的r阶矩相合性。对回归模型yi=xiβ+g(ti)+σiei(i=1,2,…,n),讨论误差为NA序列时,给出β的最小二乘估计β,研究β的r阶矩相合性。
For nonparametric regression model y_i=g(x_i)+e_i(i=1,2,...,n), the weighted kernel estimate g_n(t) for g(x) is discussed, and its r-th mean consistency under negatively associated(NA) error sequences is obtained. For y_i=x_iβ+g(t_i)+σ_ie_i(i=1,2,...,n), the least squares are used to define _n for β, and its r-th mean consistency under NA error sequences is also obtained.
出处
《合肥工业大学学报（自然科学版）》
CAS
CSCD
北大核心
2005年第4期441-445,共5页
Journal of Hefei University of Technology：Natural Science
关键词
回归模型
NA序列
r阶矩相合
regression model
negatively associated(NA) sequence
r-th mean consistency
