Parameter estimation for a class of radial basis function-based nonlinear time-series models with moving average noises *

被引：5

作者：

Zhou, Yihong ^{[1
]}

Wang, Yanjiao ^{[2
]}

Ma, Fengying ^{[3
]}

Ding, Feng ^{[1
]}

Hayat, Tasawar ^{[4
]}

机构：

[1] Jiangnan Univ, Sch Internet Things Engn, Key Lab Adv Proc Control Light Ind, Minist Educ, Wuxi 214122, Jiangsu, Peoples R China

[2] Guangdong Univ Technol, Sch Automat, Guangzhou 510006, Peoples R China

[3] Qilu Univ Technol, Shandong Acad Sci, Sch Elect Engn & Automat, Jinan 250353, Peoples R China

[4] King Abdulaziz Univ, Dept Math, Jeddah 21589, Saudi Arabia

来源：

JOURNAL OF THE FRANKLIN INSTITUTE-ENGINEERING AND APPLIED MATHEMATICS | 2021年 / 358卷 / 04期

基金：

中国国家自然科学基金;

关键词：

TRACKING CONTROL; DISTURBANCE REJECTION; IDENTIFICATION METHOD; FAULT-DIAGNOSIS; ALGORITHM; NETWORKS; OPTIMIZATION; MANIPULATOR; SELECTION; SYSTEMS;

D O I：

10.1016/j.jfranklin.2021.01.020

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

This paper focuses on the parameter estimation for radial basis function-based state-dependent autoregressive models with moving average noises (RBF-ARMA models). An extended projection algorithm is derived based on the negative gradient search. In order to reduce the sensitivity of the algorithm to noise and reduce the fluctuations of the parameter estimation errors, a modified extended stochastic gradient algorithm is proposed. By introducing a moving data window, a modified moving data window-based extended stochastic gradient algorithm is further developed to improve the parameter estimation accuracy. The simulation results show that the proposed algorithms can effectively estimate the parameters of the RBF-ARMA models. (c) 2021 The Franklin Institute. Published by Elsevier Ltd. All rights reserved.

引用

页码：2576 / 2595

页数：20

共 50 条

[31] Modelling gene expression time-series with radial basis function neural networks
Möller-Levet, CS
Yin, HJ
Cho, KH
Wolkenhatier, O
2004 IEEE INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS, VOLS 1-4, PROCEEDINGS, 2004, : 1191 - 1195
[32] PARAMETER-ESTIMATION FOR SOME TIME-SERIES MODELS WITHOUT CONTIGUITY
DAVIS, RA
ROSENBLATT, M
STATISTICS & PROBABILITY LETTERS, 1991, 11 (06) : 515 - 521
[33] Radial basis function approach to nonlinear Granger causality of time series
Ancona, N
Marinazzo, D
Stramaglia, S
PHYSICAL REVIEW E, 2004, 70 (05): : 7 - 1
[34] Online Parameter Optimization Scheme for the Kernel Function-Based Mixture Models Disturbed by Colored Noises
Li, Jimei
Ding, Feng
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS II-EXPRESS BRIEFS, 2024, 71 (08) : 3960 - 3964
[35] EMBEDDING DIMENSION ESTIMATION OF CHAOTIC TIME-SERIES USING SELF-GENERATING RADIAL BASIS FUNCTION NETWORK
KATAYAMA, R
KUWATA, K
KAJITANI, Y
WATANABE, M
FUZZY SETS AND SYSTEMS, 1995, 71 (03) : 311 - 327
[36] INTERPRETATION OF R2 IN AUTOREGRESSIVE-MOVING AVERAGE TIME-SERIES MODELS
NELSON, CR
AMERICAN STATISTICIAN, 1976, 30 (04): : 175 - 180
[37] ESTIMATION OF THE PARAMETERS OF LINEAR TIME-SERIES MODELS SUBJECT TO NONLINEAR RESTRICTIONS
NAGARAJ, NK
FULLER, WA
ANNALS OF STATISTICS, 1991, 19 (03): : 1143 - 1154
[38] A BOUND FOR ESTIMATION IN NONLINEAR TIME-SERIES MODELS BY INDEPENDENCE TESTING METHODS
FEUERVERGER, A
STATISTICS & PROBABILITY LETTERS, 1988, 6 (04) : 237 - 241
[39] A local linear radial basis function neural network for financial time-series forecasting
Nekoukar, Vahab
Beheshti, Mohammad Taghi Hamidi
APPLIED INTELLIGENCE, 2010, 33 (03) : 352 - 356
[40] A local linear radial basis function neural network for financial time-series forecasting
Vahab Nekoukar
Mohammad Taghi Hamidi Beheshti
Applied Intelligence, 2010, 33 : 352 - 356

← 1 2 3 4 5 →