Regularized estimation procedures for statistically indeterminate singular linear models

被引:0
作者
Pankov, AR [1 ]
Semenikhin, KV [1 ]
机构
[1] Moscow Inst Aviat Technol, Moscow, Russia
来源
PROCEEDINGS OF THE 41ST IEEE CONFERENCE ON DECISION AND CONTROL, VOLS 1-4 | 2002年
关键词
D O I
暂无
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
The problem of minimax estimation for the linear multivariate regression model with uncertain second-order statistics is considered. The framework of dual optimization applied recently to the regular models is developed for the singular case by means of the regularization procedure. The results obtained are illustrated by an example of a singular model.
引用
收藏
页码:2625 / 2626
页数:2
相关论文
共 8 条
[1]  
ALBERT A, 1972, REGRESSION MOOR PENR
[2]  
LIPTSER RS, 1977, STAT RANDOM PROCESSE
[3]  
Matasov A. I., 1998, ESTIMATORS UNCERTAIN
[4]  
Pankov AR, 2000, AUTOMAT REM CONTR+, V61, P785
[5]  
PANKOV AR, 1998, AUTOMATION REMOTE 2, V59, P158
[6]  
Tikhonov A. N., 1979, Methods of Solving Ill-Posed Problems
[7]   MINIMAX LINEAR OBSERVERS AND REGULATORS FOR STOCHASTIC-SYSTEMS WITH UNCERTAIN 2ND-ORDER STATISTICS [J].
VERDU, S ;
POOR, HV .
IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 1984, 29 (06) :499-511
[8]  
Verdu S., 1984, IEEE Transactions on Information Theory, VIT-30, P328, DOI 10.1109/TIT.1984.1056876