[1]贺传富,周惠新,虞献群.非线性约束条件的污染数据的L2-估计[J].东南大学学报(自然科学版),2004,34(5):686-689.[doi:10.3969/j.issn.1001-0505.2004.05.028]
 He Chuanfu,Zhou Huixin,Yu Xianqun.L 2-estimators of contaminated data with nonlinear constraints[J].Journal of Southeast University (Natural Science Edition),2004,34(5):686-689.[doi:10.3969/j.issn.1001-0505.2004.05.028]
点击复制

非线性约束条件的污染数据的L2-估计()
分享到:

《东南大学学报(自然科学版)》[ISSN:1001-0505/CN:32-1178/N]

卷:
34
期数:
2004年第5期
页码:
686-689
栏目:
数学、物理学、力学
出版日期:
2004-09-20

文章信息/Info

Title:
L 2-estimators of contaminated data with nonlinear constraints
作者:
贺传富1 周惠新2 虞献群3
1 东南大学数学系, 南京 210096; 2 南京财经大学应用数学系, 南京 210003; 3 南京市建设银行,南京 210002
Author(s):
He Chuanfu1 Zhou Huixin2 Yu Xianqun3
1 Department of Mathematics, Southeast University, Nanjing 210096, China
2 Department of Applied Mathematics, Nanjing University of Finance and Economics, Nanjing 210003, China
3 Nanjing Construction Bank, Nanjing
关键词:
L 2-估计 污染数据 非线性约束
Keywords:
L 2-estimator contaminated data nonlinear constraint
分类号:
211.64
DOI:
10.3969/j.issn.1001-0505.2004.05.028
摘要:
讨论非线性回归模型yi=f(xi,β)+ei,i=1,2,…,n,其中Eei=0,Eei221,221假设yi受到另一独立同分布随机变量序列ui的污染,仅能观察到污染数据y*i=(1-v)yi+vui,0≤v<1,v为未知参数.估计问题带有等式和不等式约束,约束是非线性的.首先利用矩估计方法给出污染参数的估计,最后利用最优化方法讨论了其非线性约束的L2-估计渐近问题,并给出估计量是概率有界的等相关结果.
Abstract:
This paper studies nonlinear regression problem yi=f(xi,β)+ei,i=1,2,…,n, where Eei=0,Ee2i21.221 Assume that yi(i=1,2,…,n) are contaminated by another i.i.d.(independent identically distributed)random variable sequence ui,and only the contaminated data y*i=(1-v)yi+vui are observable, where 0≤v<1 is unknown contaminated parameter. The estimator problem has equality and inequality constraints, and also the constraints are nonlinear. First, estimators of contaminated data are given according to the method of moment estimate. Finally, this paper studies the asymptotic problem of L2-estimator with nonlinear constraint and provides relevant conclusions that estimator is bounded in probability.

参考文献/References:

[1] Yu K F.A note on the estimation of the mixing parameter in mixture of two distributions [J].Commun Ststist, 1991,20:595-609.
[2] 郑祖康,吴雪明,饶刚.污染数据的处理[J].应用概率统计,1998,14(3):307-312.
  Zheng Zukang,Wu Xueming,Rao Gang.The treatment of contaminated data [J].Chinese Journal of Applied Probability and Statistics, 1998,14(3):307-312.(in Chinese)
[3] 陈明华.污染数据回归分析中估计的强相关性[J].应用概率统计.1998,14(1):73-78.
  Chen Minghua.Strong consistency of estimates in regression analysis for contaminattion data [J].Chinese Journal of Applied Probability and Statistics, 1998,14(1):73-78.(in Chinese)
[4] 任哲,陈明华.污染数据回归分析中参数的最小一乘估计 [J].应用概率统计,2000,16(3):262-268.
  Ren Zhe,Chen Minghua.Least absolute deviation estimator of parameter in regression analysis for contaminattion data [J].Chinese Journal of Applied Probability and Statistics,2000,16(3):262-268.(in Chinese)
[5] Shao X F.Asymptotic behavior of estimates in nonlinear regression analysis of contaminated data [J].Journal of Nanjing University Mathematical Biquarterly, 2001,19(1):1-13.
[6] Wang J.Asymptotic of least-squares estimators for constrained nonlinear regression [J]. Ann Statistics,1996,24:1316-1326.
[7] Jennrich R.Asymptotic properties of nonlinear least squares estimators [J].Ann Math Statist,1969,40:633-643.
[8] Billingsleye P. Convergence of probability measures [M].New York:John Wiley,1968.201-240.

备注/Memo

备注/Memo:
作者简介: 贺传富(1965—),男,硕士,讲师,he1965fu@163.com.
更新日期/Last Update: 2004-09-20