Scientific Journal

Herald of Advanced Information Technology

NON-STATIONARY TIME SERIES PREDICTION USING ONE-DIMENSIONAL CONVOLUTIONAL NEURAL NETWORK MODELS
Abstract:
The main goal of non-stationary time series prediction is the construction, identification, configuration and verification of their models. The efficiency of using machine learning technologies for the analysis of non-stationary time series is shown due to their ability to model complex nonlinear dependencies in the behaviour of the time series from both depending on previous values and external factors, to analyse features, relationships and complex interactions. The features of time series prediction using a one-dimensional convolutional neural network are discussed. The features of the architecture and the training process when using a one-dimensional convolutional neural network are considered on the example of solving the problems to predict sales and build a forecast of company stock prices. To improve the quality of the prediction, the initial time series were preprocessed by the moving average method in the window. Computer modelling of the predicting problem using the one-dimensional convolutional neural network was performed in the Python programming language. The sales prediction using the proposed onedimensional convolutional neural network model predicted volume sale of cars and commercial vehicles in Vietnam from two thousand and eleven to two thousand and eighteen. The one-dimensional convolutional neural network model has given a high accuracy of prediction with seasonal trend data. In stock prices prediction, another architecture of one-dimensional convolutional neural network model was launched, which corresponds to non-stationary data with large lengths of data series with small intervals between minutes, such as stock trading statistics per minute. In this project, data is taken from Amazon Nasdaq one hundred for forty thousand five hundred and sixty data points. The data is divided into training and test sets. The test suite is used to verify the actual performance of the model. It is shown that the model of a one-dimensional convolutional neural network gives good results in the presence of both seasonal and trend components of the time series with large data sizes
Authors:
Keywords
DOI
10.15276/hait01.2020.3
References
1. Alexander, A. (2014). “Case studies: business intelligence”. Accounting Today 28(6), p.32. Google Scholar. “Towards Differentiating Business Intelligence, Big Data, Data Analytics and Knowledge Discovery”. [Digital Resource]. – Access mode: https://www.accountingtoday.com/, Title from the screen. – Active link – 10.11.19. 
2. Hyndman, R. & Athanasopoulos, G. (2017). “Forecasting: Principles and Practice”. (2nded.). [Digital Resource]. – Access mode: https://ww.otexts.org/fpp2, Title from the screen. – Active link – 12.12.19. Herald of Advanced Information Technology 2020; Vol. 3 No.1: 362–372 Intellectual Information Technologies: Neural Networks, Machine Learning, Forecasting ISSN 2663-0176 (Print) 369 ISSN 2663-7731 (Online) 
3. Roman Josue, D. (2018). “7 Ways Time Series Forecasting Differs from Machine Learning”. [Digital Resource]. – Access mode: https://www.datascience.com/blog/ time-seriesforecasting-machine-learning-differences. Title from the screen. – Active link – 5.11.19.
4. Heschmeyer, A. (2019). “Forecasting Walmart Sales Using Machine Learning Models”. [Digital Resource]. – Access mode: https://medium.com/@auggieheschmeyer/forecastin g -walmart-sales-using-machine-learning-models3bf38f6c533. Title from the screen. – Active link – 14.06.19. 
5. Gegic, E., Isakovic, B., Keco, D., Masetic, Z. & Kevric, J. (2019). Car Price “Prediction using Machine Learning Techniques”. TEM Journal. Volume 8, Issue 1, pp. 113-118. DOI: 10.18421/TEM81-16. 6. Zhong, X. & Enke, D. (2019). “Predicting the daily return direction of the stock market using hybrid machine learning algorithms”. Financial Innovation. Volume 5, Issue 24, pp. 1-20. DOI: https:// doi.org/10.1186/s40854-019-0138-0. 
7. Hailong, C., Rajagopalan S. & Ward A. (June 2019). “Predicting product return volume using machine learning methods”. European Journal of Operational Research. Vol. 281, Issue 3, pp. 612- 627. DOI: https://doi.org/10.1016/ j.ejor.2019.05.046. 8. Potdar, J. & Mathew, R. (2018). “Machine Learning Algorithms in Stock Market Prediction”. In: Pandian A., Senjyu T., Islam S., Wang H. (eds.). Proceeding of the International Conference on Computer Networks, Big Data and IoT (ICCBI - 2018). ICCBI 2018. Lecture Notes on Data Engineering and Communications Technologies. Vol. 31. Springer, Cham. DOI: https: doi.org/10.1007/978-3-030-24643-3_23. 
9. Chollet, F. (December, 2018). “Deep learning with Python”. Manning Publications Co., USA. 1 st Edition, 384 p. 
10. Tymchenko, B., Hramatik, A., Tulchiy, H., Antoshchuk, S. (2019) “Classifying mixed patterns of proteinsin microscopic images with deep neural networks”. Herald of Advanced Information Technology, 2019, Vol.2, No.1, pp. 29-36. DOI: 10.15276/hait.01.2019.3. 
11. Arsirii, O., Antoshchuk,S., Babilunha, O. & Manikaeva, O. (2019) “Intellectual Information Technology of Analysis of Weakly-Structured MultiDimensional Data of Sociological Research”. Material of International Scientific Conference “Intellectual Systems of Decision Making and Problem of Computational Intelligence”, pp. 242-258. 12. Goodfellow, I., Yoshua, Y. & Courville, A. (November, 2016). “Deep Learning (Adaptive Computation and Machine Learning series)”. The MIT Press, 800 p. 
13. Mai, F., Tian, S., Lee, C. & Ma, L. (April 2019). “Deep Learning Models for Bankruptcy Prediction using Textual Disclosures”. European Journal of Operational Research. Volume 274, Issue 2, p. 743-758. DOI https://doi.org/10.1016/j.ejor.2018.10.024. 
14. Jain, S., Gupta, R. & Moghe, A. (2018) “Stock Price Prediction on Daily Stock Data using Deep Neural Networks”. International Journal of Neural Networks and Advanced Applications. Volume 5, 2018. [Digital Resource]. – Access mode: http://www.naun.org/ main/NAUN/neural/ 2018/a162016-062.pdf. Title from the screen. – Active link – 17.07.18. 
15. Boris, B. (Jan, 2019). “Using the latest advancements in deep learning to predict stock price movements”. Towards Data Science. AI Quant in Hong Kong. MSc Data Science @ UC Berkeley. MSc Finance @ MBS. [Digital Resource]. – Access mode: https://towardsdatascience.com/ aifortrading2edd6fac689d. Title from the screen. – Active link – 19.01.19. 
16. Sadouk, L. (November, 2018). “CNN Approaches for Time Series classification”. Published: November 5th 2018. DOI: 10.5772/intechopen.81170. 
17. Machine Learning Blogby owygs156 (May, 2017). “Forecasting Time Series Data with Convolutional Neural Networks”. [Digital Resource]. – Access mode: https://intelligentonlinetools.com/blog/2017/05/14/ti me-series-prediction-with-convolutional-neuralnetworks/. 
18. Jon, C. (2018). “Forecasting Time Series Data with Convolutional Neural Networks”. [Digital Resource]. – Access mode: https://medium.com/@jon.froiland/ convolutionalneural-networks-for-sequence-processing-part-1- 420dd9b500.Title from the screen. – Active link – .09.18. 
19. Kondek, J., Chennoufi, K. & Noessler, K. (2019). “Convolutional neural networks – Sales forecast”. Seminar information systems (WS18/19) | February 8, 2019. [Digital Resource]. – Access mode: https://humboldt-wi.github.io/blog/research/ information_systems_1819/cnn_sales_forecast/. Title from the screen. – Active link – .02.19. 
20. Cheriyan, S., Ibrahim, S., Mohanan, S. & Treesa, S. (2018). “Intelligent Sales Prediction Using Machine Learning Techniques”. International Conference on Computing, Electronics & Herald of Advanced Information Technology 2020; Vol. 3 No.1: 362–372 Intellectual Information Technologies: Neural Networks, Machine Learning, Forecasting 370 ISSN 2663-0176 (Print) ISSN 2663-7731 (Online) Communications Electronic 2018. DOI: 10.1109/iCCECOME.2018. 8659115. 
21. Chen, S. & He, H. (2018). “Stock Prediction Using Convolutional Neural Network”. AIAAT 2018. IOP Publishing. IOP Conf. Series: Materials Science and Engineering, 435. DOI:10.1088/1757-899X/435/1/012026. 
22. Brownlee, J. (2017). “Python Environment for Time Series Forecasting”. [Digital Resource]. – Access mode: https://machinelearningmastery.com/pythonenvironment-time-series-forecasting. Title from the screen. – Active link – 8.03.17. 
23. (2019) “Scikit-learn. Machine Learning in Python. Model evaluation: quantifying the quality of predictions”. [Digital Resource]. – Access mode: https://scikit-learn.org/stable/modules/model_ evaluation.html. Title from the screen. – Active link – 3.07.19. 
24. Ackermann, N. (September, 2019). “Introduction to 1D Convolutional Neural Networks in Keras for Time Sequences”. Good Audience. The Front Page of Deep Tech. [Digital Resource]. – Access mode: https://blog.goodaudience.com/ introduction-to-1d-convolutional-neural-networksin-keras-for-time-sequences-3a7ff801a2cf.Title from the screen. – Active link – 4.09.19. 
25. (2019) “Keras Documentation. Keras: the Python Deep Learning library”. [Digital Resource]. – Access mode: https://keras.io. Title from the screen. – Active link – 24.10.19. 
26. Brownlee, J. (2018). “How to Develop Convolutional Neural Network Models for Time Series Forecasting. Deep Learning for Time Series”. Active link: November 12, 2018. [Digital Resource]. – Access mode: https://machine learningmastery.com/how-to-develop-convolutionalneural-network-models-for-time-series-forecasting. Title from the screen. – Active link – 5.11.18. 
27. (2018). Internet portal VERcity. “Sales of cars and commercial vehicles in Vietnam” [in Russian]. [Digital Resource]. – Access mode: https://auto.vercity.ru/statistics/sales/asia/vietnam. Title from the screen. – Active link – 30.10.18. 
28. (2017) Department of Computer Science and Engineering. “NASDAQ 100 stock data”. University of California, San Diego, USA. [Digital Resource]. – Access mode: http://cseweb.ucsd.edu/~ yaq007/NASDAQ100_stock_data.html. Title from the screen. – Active link – 1.04.17. 
29. Kim, T. & Kim, H. Y. (February, 2019). “Forecasting stock prices with a feature fusion LSTM-CNN model using different representations of the same data”. PLOS ONE Journal. Vol. 14(2). DOI: https://doi.org/10.1371/ journal.pone.0212320.

Received. 30.11.2019  
Received after revision 27.12.2019 
Accepted 18.01. 2020
Published:
Last download:
24 Oct 2021

Contents


[ © KarelWintersky ] [ All articles ] [ All authors ]
[ © Odessa National Polytechnic University, 2018.]