摘要
在α-混合误差下讨论Priestly和Chao提出的一类非参数回归函数加权核估计的相合性.对文献[1]中,{ε_i}为强混合误差序列的情形,添加了不同的条件及计算得到g_n{x}→g(x),a,s的结论.
In this paper, we discuss the consistency of weighted kernel estimator of nonparametric regression function under a-mixing errors by Priestley and Chao. In references [1],{ε_i} is a-mixing error sequence. The results that g(x) is estimated with g_n(x) is obtained under the different conditions and calculation.
引文
[1]于德明.α—混合误差下回归函数加权和估计的强相合性[J].浙江工业大学,2006,34(4):470-472.
[2]BOENTE,G and FRAIMAN, R.Consistency of a nonparametric estimate of a density function for dependent variables[J].JOURNAL OF MULTIVAIATE ANALYSIS,25(1988):90-99.
[3]Priestley,M.B.and Chao,M.T.Nonparametric function fitting[J].J.R. Statist. Soc. B,34(1972):385-392.
[4]Benedtti,J.K. On the nonparametric estimati on of regression functions[J].J.R.statist.soc.B39(1977):248-253.
[5]秦永松.非参数回归函数估计得一个结果[J].工程数学学报,1989,6(3):120-123.
[6]杨善朝.φ-混合误差下非参数回归函数估计的强相合性[J].高校应用数学学报,1995,10(2):173—179.
[7]杨善朝,王岳宝.NA样本回归函数估计的强相合性[J].应用数学学报,1999,22(4):522—530.
[8]李乃医,黄娟.p-混合误差下非参数回归函数加权核估计的相合性[J].江西科学.2006,24(4):147—148.
[9]陆传荣,林正炎.混合相依变量的极限理论[M].北京:科学出版社,1997:5.