×

zbMATH — the first resource for mathematics

An adaptive orthogonal search algorithm for model subset selection and non-linear system identification. (English) Zbl 1152.93343
Summary: A new adaptive orthogonal search (AOS) algorithm is proposed for model subset selection and non-linear system identification. Model structure detection is a key step in any system identification problem. This consists of selecting significant model terms from a redundant dictionary of candidate model terms, and determining the model complexity (model length or model size). The final objective is to produce a parsimonious model that can well capture the inherent dynamics of the underlying system. In the new AOS algorithm, a modified generalized cross-validation criterion, called the adjustable prediction error sum of squares (APRESS), is introduced and incorporated into a forward orthogonal search procedure. The main advantage of the new AOS algorithm is that the mechanism is simple and the implementation is direct and easy, and more importantly it can produce efficient model subsets for most non-linear identification problems.

MSC:
93B30 System identification
93B15 Realizations from input-output data
93B40 Computational methods in systems theory (MSC2010)
PDF BibTeX XML Cite
Full Text: DOI
References:
[1] DOI: 10.2307/1267500 · Zbl 0286.62044
[2] DOI: 10.1214/aos/1176347964
[3] Billings SA, Neural Network Systems Techniques and Applications pp 231– (1998)
[4] DOI: 10.1080/00207729208949363 · Zbl 0761.93077
[5] DOI: 10.1080/00207178608933633 · Zbl 0597.93058
[6] DOI: 10.1109/TNN.2005.849842
[7] Billings SA, Int. J. Cont. 49 pp 2157– (1989)
[8] Box GEP, Time Series Analysis: Forecasting and Cont., 3. ed. (1994)
[9] DOI: 10.1080/00207178908953472 · Zbl 0686.93093
[10] DOI: 10.1109/72.80341
[11] DOI: 10.1109/TSMCB.2004.828199
[12] DOI: 10.1109/TSMCB.2003.817107
[13] DOI: 10.1007/BF01404567 · Zbl 0377.65007
[14] DOI: 10.1214/aos/1176347963 · Zbl 0765.62064
[15] DOI: 10.2307/1270359 · Zbl 0672.65119
[16] DOI: 10.2307/1268518 · Zbl 0461.62059
[17] DOI: 10.1016/S0167-7152(02)00056-1 · Zbl 0996.62036
[18] DOI: 10.1109/TNN.2003.809422
[19] DOI: 10.1049/ip-cta:20040693
[20] DOI: 10.1111/j.1369-7412.2004.05500.x · Zbl 1062.62185
[21] Ljung L, System Identification: Theory for the User (1987)
[22] Miller AJ, Subset Selection in Regression (1990)
[23] Myers R, Classical and Modern Regression with Applications, 2. ed. (1990)
[24] Söderström T, System Identification (1989)
[25] DOI: 10.1080/00207178608933575 · Zbl 0605.93050
[26] Stone M, J. Roy. Statist. Soc. 36 pp 111– (1974)
[27] DOI: 10.2307/2291003 · Zbl 0813.62037
[28] DOI: 10.1111/1467-9892.00193 · Zbl 0974.62077
[29] DOI: 10.1080/02331889508802499 · Zbl 0836.62067
[30] DOI: 10.1080/00207170310001639640 · Zbl 1050.93506
[31] DOI: 10.1080/00207179608921662 · Zbl 0858.93069
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.