Two ways towards combining Sequential Neural Network and Statistical Methods to Improve the Prediction of Time Series

Statistic modeling and data-driven learning are the two vital fields that attract many attentions. Statistic models intend to capture and interpret the relationships among variables, while data-based learning attempt to extract information directly from the data without pre-processing through complex models. Given the extensive studies in both fields, a subtle issue is how to properly integrate data based methods with existing knowledge or models. In this paper, based on the time series data, we propose two different directions to integrate the two, a decomposition-based method and a method exploiting the statistic extraction of data features. The first one decomposes the data into linear stable, nonlinear stable and unstable parts, where suitable statistical models are used for the linear stable and nonlinear stable parts while the appropriate machine learning tools are used for the unstable parts. The second one applies statistic models to extract statistics features of data and feed them as additional inputs into the machine learning platform for training. The most critical and challenging thing is how to determine and extract the valuable information from mathematical or statistical models to boost the performance of machine learning algorithms. We evaluate the proposal using time series data with varying degrees of stability. Performance results show that both methods can outperform existing schemes that use models and learning separately, and the improvements can be over 60%. Both our proposed methods are promising in bridging the gap between model-based and data-driven schemes and integrating the two to provide an overall higher learning performance.

[1]  Richard A. Davis,et al.  Introduction to time series and forecasting , 1998 .

[2]  Yukika Awazu,et al.  Vagaries of the Euro: an Introduction to ARIMA Modeling , 2014 .

[3]  Alper Tokgöz,et al.  A RNN based time series approach for forecasting turkish electricity load , 2018, 2018 26th Signal Processing and Communications Applications Conference (SIU).

[4]  Matteo Sangiorgio,et al.  Robustness of LSTM neural networks for multi-step forecasting of chaotic time series , 2020 .

[5]  Alaa El. Sagheer,et al.  Time series forecasting of petroleum production using deep LSTM recurrent networks , 2019, Neurocomputing.

[6]  C. Bergmeir,et al.  Recurrent Neural Networks for Time Series Forecasting: Current Status and Future Directions , 2019, International Journal of Forecasting.

[7]  Ku Ruhana Ku-Mahamud,et al.  SWGARCH model for time series forecasting , 2017, IML.

[8]  Jason Poulos.  RNN-based counterfactual time-series prediction , 2018 .

[9]  Robert F. Engle,et al.  ARCH: Selected Readings , 1995 .

[10]  Dr.N.M.Elango K.K.Sureshkumar,et al.  Performance Analysis of Stock Price Prediction using Artificial Neural Network , 2012 .

[11]  Ani Shabri,et al.  Modelling and forecasting monthly crude oil price of Pakistan: A comparative study of ARIMA, GARCH and ARIMA Kalman model , 2016 .

[12]  Liao Zhu.  The Adaptive Multi-Factor Model and the Financial Market , 2021, ArXiv.

[13]  J. Zakoian,et al.  GARCH Models: Structure, Statistical Inference and Financial Applications , 2010 .

[14]  Lovekesh Vig,et al.  ConvTimeNet: A Pre-trained Deep Convolutional Neural Network for Time Series Classification , 2019, 2019 International Joint Conference on Neural Networks (IJCNN).

[15]  Junqiang Wang,et al.  Time-series well performance prediction based on Long Short-Term Memory (LSTM) neural network model , 2020 .

[16]  Martin T. Wells,et al.  A News-based Machine Learning Model for Adaptive Asset Pricing , 2021, ArXiv.

[17]  High-Dimensional Estimation, Basis Assets, and the Adaptive Multi-Factor Model , 2018 .

[18]  R. Shanmugam.  Introduction to Time Series and Forecasting , 1997 .

[19]  Rohit Bhakar,et al.  Very Short term Wind Power Prediction Using Hybrid Univariate ARIMA-GARCH Model , 2019, 2019 8th International Conference on Power Systems (ICPS).

[20]  Navin Kumar Manaswi,et al.  RNN and LSTM , 2018 .

[21]  Luca Di Persio,et al.  Artificial Neural Networks architectures for stock price prediction: comparisons and applications , 2016 .

[22]  Aderemi Oluyinka Adewumi,et al.  Stock Price Prediction Using the ARIMA Model , 2014, 2014 UKSim-AMSS 16th International Conference on Computer Modelling and Simulation.