FROM TIME-SERIES TO LINEAR-SYSTEM .3. APPROXIMATE MODELING

被引：161
作者：
WILLEMS, JC
机构：
[1] Univ of Groningen, Groningen, Neth, Univ of Groningen, Groningen, Neth
来源：
AUTOMATICA | 1987年 / 23卷 / 01期
关键词：
DATA PROCESSING - Data Reduction and Analysis - MATHEMATICAL TECHNIQUES - Correlation Methods;
D O I：
10.1016/0005-1098(87)90120-8
中图分类号：
TP [自动化技术、计算机技术];
学科分类号：
0812 ;
摘要：
An optimal approximate model, defined in terms of the complexity of a model and the misfit between a model and the observed data, yields algorithms for computing an optimal fitting model with a maximal admissible complexity or, alternatively, a minimally complex model which explains an observed time series up to a maximally tolerated misfit.
引用
收藏
页码：87 / 115
页数：29
相关论文
共 11 条
  • [1] Akaike H., 1977, APPL STAT, V543, P27
  • [2] BEKKER P, 1986, IN PRESS PSYCHOMETRI
  • [3] Caines P. E., 1986, Modelling, Identification and Robust Control, P563
  • [4] DEISTLER M, 1986, SPRINGER LECTURE NOT, V86, P37
  • [5] WE CAN DO SOMETHING ABOUT MULTICOLLINEARITY
    KALMAN, RE
    [J]. COMMUNICATIONS IN STATISTICS-THEORY AND METHODS, 1984, 13 (02) : 115 - 125
  • [6] KOLMOGOROV A, 1965, PROBL PEREDACHI INF, V1, P4
  • [7] Ljung L., 1987, SYSTEM IDENTIFICATIO
  • [8] Ljung L, 1983, THEORY PRACTICE RECU
  • [9] LJUNG L, 1985, 24TH P IEEE C DEC CO, P1056
  • [10] RISSANEN J, 1985, UNPUB IBM RES K54282
12