Based on the work in Ding and Ding(2008),we develop a modifed stochastic gradient(SG)parameter estimation algorithm for a dual-rate Box-Jenkins model by using an auxiliary model.We simplify the complex dual-rate Box-J...Based on the work in Ding and Ding(2008),we develop a modifed stochastic gradient(SG)parameter estimation algorithm for a dual-rate Box-Jenkins model by using an auxiliary model.We simplify the complex dual-rate Box-Jenkins model to two fnite impulse response(FIR)models,present an auxiliary model to estimate the missing outputs and the unknown noise variables,and compute all the unknown parameters of the system with colored noises.Simulation results indicate that the proposed method is efective.展开更多
Artificial neural network models are a popular estimation tool for fitting nonlinear relationships because they require no assumptions about the form of the fitting function,non-Gaussian distributions,multicollinearit...Artificial neural network models are a popular estimation tool for fitting nonlinear relationships because they require no assumptions about the form of the fitting function,non-Gaussian distributions,multicollinearity,outliers and noise in the data.The problems of backpropagation models using artificial neural networks include determination of the structure of the network and overlearning courses.According to data from 1981 to 2008 from 15 permanent sample plots on Dagangshan Mountain in Jiangxi Province,a back-propagation artificial neural network model(BPANN)and a support vector machine model(SVM)for basal area of Chinese fir(Cunninghamia lanceolata)plantations were constructed using four kinds of prediction factors,including stand age,site index,surviving stem numbers and quadratic mean diameters.Artificial intelligence methods,especially SVM,could be effective in describing stand basal area growth of Chinese fir under different growth conditions with higher simulation precision than traditional regression models.SVM and the Chapman–Richards nonlinear mixed-effects model had less systematic bias than the BPANN.展开更多
基金Project supported by the National Natural Science Foundation of China(No.60973043)the Natural Science Foundation of Jiangsu Province,China(No.BK20131109)
文摘Based on the work in Ding and Ding(2008),we develop a modifed stochastic gradient(SG)parameter estimation algorithm for a dual-rate Box-Jenkins model by using an auxiliary model.We simplify the complex dual-rate Box-Jenkins model to two fnite impulse response(FIR)models,present an auxiliary model to estimate the missing outputs and the unknown noise variables,and compute all the unknown parameters of the system with colored noises.Simulation results indicate that the proposed method is efective.
基金supported by the National Scientific and Technological Task in China(Nos.2015BAD09B0101,2016YFD0600302)National Natural Science Foundation of China(No.31570619)the Special Science and Technology Innovation in Jiangxi Province(No.201702)
文摘Artificial neural network models are a popular estimation tool for fitting nonlinear relationships because they require no assumptions about the form of the fitting function,non-Gaussian distributions,multicollinearity,outliers and noise in the data.The problems of backpropagation models using artificial neural networks include determination of the structure of the network and overlearning courses.According to data from 1981 to 2008 from 15 permanent sample plots on Dagangshan Mountain in Jiangxi Province,a back-propagation artificial neural network model(BPANN)and a support vector machine model(SVM)for basal area of Chinese fir(Cunninghamia lanceolata)plantations were constructed using four kinds of prediction factors,including stand age,site index,surviving stem numbers and quadratic mean diameters.Artificial intelligence methods,especially SVM,could be effective in describing stand basal area growth of Chinese fir under different growth conditions with higher simulation precision than traditional regression models.SVM and the Chapman–Richards nonlinear mixed-effects model had less systematic bias than the BPANN.
