基于辅助粒子滤波与灰色预测的时间序列NAR模型状态估计
|
摘要
在非线性自回归(NAR)模型建模的基础上,文章利用辅助粒子滤波(APF)和灰色预测(GM)相结合的方法估计NAR模型的参数和状态,减少因参数估计问题带来的状态估计误差。并将其与传统NAR模型估计和基于粒子滤波估计NAR模型状态的方法进行实验对比。结果表明,基于辅助粒子滤波与灰色预测相结合的估计方法优于传统NAR模型和粒子滤波估计方法,更适合于金融时间序列的预测。
Based on the modeling of nonlinear autoregressive(NAR) model, this paper uses the method of combining the auxiliary particle filter(APP) and grey prediction(GM) to estimate the parameter and state of NAR model and reduce the state estimation error caused by parameter estimation problem. The paper also makes an experimental comparison between the proposed method with the traditional NAR model estimation method and NAR model state estimation based on particle filter. The results show that the estimation method based on the combination of auxiliary particle filter and grey prediction is better than the traditional NAR model and particle filter, and more suitable for the prediction of financial time series.
Based on the modeling of nonlinear autoregressive(NAR) model, this paper uses the method of combining the auxiliary particle filter(APP) and grey prediction(GM) to estimate the parameter and state of NAR model and reduce the state estimation error caused by parameter estimation problem. The paper also makes an experimental comparison between the proposed method with the traditional NAR model estimation method and NAR model state estimation based on particle filter. The results show that the estimation method based on the combination of auxiliary particle filter and grey prediction is better than the traditional NAR model and particle filter, and more suitable for the prediction of financial time series.
引文
[1]Yan Y S,Heng J C,Kai C L.Parameter Estimation of Varying Coefficients Structural EV Model With Time Series[J].Acta Mathematica Sinica,English Series,2017,33(5).
[2]Huang K,Ma J X,Zhang Y h,et al.Nonlinear Time Series Model Identification and Its Application in Structural Damage Detection[C].Proceedings of 2016 IEEE Information Technology,Networking,Electronic and Automation Control Conference,2016.
[3]Lopes H F,Tsay R S.Particle Filters and Bayesian Inference In Financial Econometrics[J].Journal of Forecasting,2011,30(1).
[4]Djuric P M,Khan M,Johnston D E.Particle Filtering of Stochastic Volatility Modeled With Leverage[J].IEEE Journal of Selected Topics in Signal Processing,2012,(6).
[5]Sheinson D M,Niem I J,Meiring W.Comparison of the Performance of Particle Filter Algorithms Applied to Tracking of a Disease Epidemic[J].Mathematical Biosciences,2014,255(3).
[6]Yassin I M,Zabidi A,Salleh M K M,et al.Malaysian Tourism Interest Forecasting Using Nonlinear Auto-regressive(NAR)Model[C].2013 IEEE 3rd International Conference on System Engineering and Technology,2013.
[7]Amisigo B A,Giesen V D N,Rogers C,et al.Monthly Streamflow Prediction in the Volta Basin of West Africa:A SISO NARMAX polynomial Modelling[J].Physics and Chemistry of the Earth,2007,(33).
[8]陈亚静,蔡如华,桂丛楠.基于粒子滤波的股价预测方法[J].统计与决策,2017,(3).
[9]Yuan Y,Lu W.An Auxiliary Particle Filter for Nonlinear Dynamic Equilibrium Models[J].Economics Letters,2016,144(10).
[10]Yohan P,Franois D.Optimal SIR Algorithm vs.Fully Adapted Auxiliary Particle Filter:A Non Asymptotic Analysis[J].Statistics and Computing,2013,23(6).
[11]Chen J F,Shi Z G,Hong S H,et al.Grey Prediction Based Particle Filterfor Maneuvering Target Tracking[J].Progress in Electromagnetic Research,2009.
[12]朱明清,王智灵,陈宗海.基于灰色预测模型和粒子滤波的视觉目标跟踪算法[J].控制与决策,2012,27(1).
[2]Huang K,Ma J X,Zhang Y h,et al.Nonlinear Time Series Model Identification and Its Application in Structural Damage Detection[C].Proceedings of 2016 IEEE Information Technology,Networking,Electronic and Automation Control Conference,2016.
[3]Lopes H F,Tsay R S.Particle Filters and Bayesian Inference In Financial Econometrics[J].Journal of Forecasting,2011,30(1).
[4]Djuric P M,Khan M,Johnston D E.Particle Filtering of Stochastic Volatility Modeled With Leverage[J].IEEE Journal of Selected Topics in Signal Processing,2012,(6).
[5]Sheinson D M,Niem I J,Meiring W.Comparison of the Performance of Particle Filter Algorithms Applied to Tracking of a Disease Epidemic[J].Mathematical Biosciences,2014,255(3).
[6]Yassin I M,Zabidi A,Salleh M K M,et al.Malaysian Tourism Interest Forecasting Using Nonlinear Auto-regressive(NAR)Model[C].2013 IEEE 3rd International Conference on System Engineering and Technology,2013.
[7]Amisigo B A,Giesen V D N,Rogers C,et al.Monthly Streamflow Prediction in the Volta Basin of West Africa:A SISO NARMAX polynomial Modelling[J].Physics and Chemistry of the Earth,2007,(33).
[8]陈亚静,蔡如华,桂丛楠.基于粒子滤波的股价预测方法[J].统计与决策,2017,(3).
[9]Yuan Y,Lu W.An Auxiliary Particle Filter for Nonlinear Dynamic Equilibrium Models[J].Economics Letters,2016,144(10).
[10]Yohan P,Franois D.Optimal SIR Algorithm vs.Fully Adapted Auxiliary Particle Filter:A Non Asymptotic Analysis[J].Statistics and Computing,2013,23(6).
[11]Chen J F,Shi Z G,Hong S H,et al.Grey Prediction Based Particle Filterfor Maneuvering Target Tracking[J].Progress in Electromagnetic Research,2009.
[12]朱明清,王智灵,陈宗海.基于灰色预测模型和粒子滤波的视觉目标跟踪算法[J].控制与决策,2012,27(1).