Information technology for time series forecasting by the method of the forecast scheme synthesis

2021;
: 81-86
https://doi.org/10.23939/ujit2021.02.081
Received: September 29, 2021
Accepted: November 23, 2021
1
Uzhhorod National University, Uzhhorod, Ukraine
2
Uzhhorod National University, Uzhhorod, Ukraine
3
Lviv Polytechnic National University, Lviv, Ukraine
4
Uzhhorod National University, Uzhhorod, Ukraine

The study is devoted to the development of information technology for forecasting based on time series. It has been found that it is important to develop new models and forecasting methods to improve the quality of the forecast. Information technology is based on the evolutionary method of synthesis of the forecast scheme grounded on basic forecast models. The selected method allows you to consider any number of predictive models that may belong to different classes. For a given time series, the weight coefficients with which the models are included in the resulting forecast scheme are calculated by finding the solution to the optimization problem. The method of constructing the objective function for the optimization problem in the form of a linear combination of forecasting results by basic forecasting models is shown. It is proposed to find the solution to the optimization problem using a genetic algorithm. The result of the method is the forecast scheme, which is a linear combination of basic forecast models. To assess the quality of the forecast, it is suggested to use forecasting errors or forecast volatility calculated as the standard deviation. Forecast quality criteria are selected depending on the context of the task. The use of forecast volatility as a quality criterion, with repeated use of technology, will reduce the deviation of forecast values from real data. The structural scheme of information technology is developed. Structurally, information technology consists of two blocks: data processing and interpretation of the obtained values. The result of the application of the developed information technology is the production rules for determining the predicted value of the studied quantity. Experimental verification of the obtained results was performed. The problem of forecasting the number of religious organizations in Ukraine based on statistical data from 1997 to 2000 has been solved. The autoregression method and the linear regression model were chosen as the basic forecast models. Based on the results of using the developed information technology, the weights of the basic models were calculated. It is demonstrated that the obtained forecast scheme allowed to improve the average absolute percentage error and forecast volatility in comparison with the selected models.

  1. Biloshchytskyi, A., Biloshchytska, S., Kuchansky, A., Bielova, O., & Andrashko, Y. (2018, February). Infocommunication system of scientific activity management on the basis of project-vector methodology. In 2018 14th International Conference on Advanced Trends in Radioelecrtronics, Telecommunications and Computer Engineering (TCSET), 200–203. IEEE. https://doi.org/10.1109/TCSET.2018.8336186
  2. Brockwell, P. J., Brockwell, P. J., Davis, R. A., & Davis, R. A. (2016). Introduction to time series and forecasting, Springer, 434. https://doi.org/10.1007/978-3-319-29854-2
  3. Cai, Q., Zhang, D., Zheng, W., & Leung, S. C. (2015). A new fuzzy time series forecasting model combined with ant colony optimization and auto-regression. Knowledge-Based Systems, 74, 61–68. https://doi.org/10.1016/j.knosys.2014.11.003
  4. Daradkeh, Y. I., Kirichenko, L., & Radivilova, T. (2018). Development of QoS methods in the information networks with fractal traffic. International Journal of Electronics and Telecommunications, 64, 27–32.
  5. Dolgikh, S. (2019). Categorized representations and general learning. In International Conference on Theory and Application of Soft Computing, Computing with Words and Perceptions, August 2019, Springer, Cham, 93–100. https://doi.org/10.1007/978-3-030-35249-3_11
  6. Dolgikh, S., & Mulesa, O. (2021) Covid-19 epidemiological factor analysis: Identifying principal factors with machine learning. CEUR Workshop Proceedings, 2833, 114–123. https://doi.org/10.1101/2020.06.01.20119560
  7. Ethnopolitics. (2020). State Service of Ukraine for Ethnopolitics and Freedom of Conscience. Retrieved from: https://dess.gov.ua/statistics-2020/
  8. Geche, F., Batyuk, A., Mulesa, O., & Voloshchuk, V. (2020, August). The Combined Time Series Forecasting Model. In 2020 IEEE Third International Conference on Data Stream Mining & Processing (DSMP), 272–275, IEEE. https://doi.org/10.1109/DSMP47368.2020.9204311
  9. Geche, F., Mulesa, O., & Buchok, V. (2017). Synthesis of generalized neural elements by means of the tolerance matrices. Eastern-European Journal of Enterprise Technologies, 4(4), 50–62. https://doi.org/10.15587/1729-4061.2017.108404
  10. Gulyanitsky, L. F., & Bondar, T. G. (2018). Research of efficiency of adaptive forecasting methods. Computer Mathematics, (1), 53–60.
  11. Hnatiienko, H., Kudin, V., Onyshchenko, A., Snytyuk, V., & Kruhlov, A. (2020, October). Greenhouse Gas Emission Determination Based on the Pseudo-Base Matrix Method for Environmental Pollution Quotas Between Countries Allocation Problem. In 2020 IEEE 2nd International Conference on System Analysis & Intelligent Computing (SAIC), 1–8. IEEE. https://doi.org/10.1109/SAIC51296.2020.9239125
  12. Hnatiienko, H., Tmienova, N., & Kruglov, A. (2021) Methods for Determining the Group Ranking of Alternatives for Incomplete Expert Rankings. In: Shkarlet S., Morozov A., Palagin A. (eds) Mathematical Modeling and Simulation of Systems (MODS2020). MODS 2020. Advances in Intelligent Systems and Computing, 1265, 217–226, Springer, Cham. https://doi.org/10.1007/978-3-030-58124-4_21
  13. Hunk, D., Rights, A. D., & Dean, W. (2003). Business forecasting, Williams.
  14. Khomytska, I., Teslyuk, V., Kryvinska, N., & Bazylevych, I. (2020). Software-based approach towards automated authorship acknowledgement – Chi-square test on one consonant group. Electronics, 9(7). https://doi.org/10.3390/electronics9071138
  15. Kirichenko, L., Radivilova, T., Bulakh, V., Zinchenko, P., & Alghawli, A. S. (2020, August). Two Approaches to Machine Learning Classification of Time Series Based on Recurrence Plots. In 2020 IEEE Third International Conference on Data Stream Mining & Processing (DSMP), August 2020, IEEE, 84–89. https://doi.org/10.1109/DSMP47368.2020.9204021
  16. Kuchansky, A., Biloshchytskyi, A., Bronin, S., Biloshchytska, S., & Andrashko, Y. (2019, October). Use of the Fractal Analysis of Non-stationary Time Series in Mobile Foreign Exchange Trading for M-Learning. In Interactive Mobile Communication, Technologies and Learning (pp. 950–961). Springer, Cham. https://doi.org/10.1007/978-3-030-49932-7_88
  17. Lupei, M., Mitsa, A., Povkhan, I., & Sharkan, V. (2020). Determining the Eligibility of Candidates for a Vacancy Using Artificial Neural Networks. In 2020 IEEE Third International Conference on Data Stream Mining & Processing (DSMP), August 2020. IEEE, 18–22. https://doi.org/10.1109/DSMP47368.2020.9204020.
  18. Mulesa, O. Yu. (2015). Design features of the information technology for the labor migrants group structure determination. Eastern-European Journal of Enterprise Technologies, 4(2), 4–8. https://doi.org/10.15587/1729-4061.2015.47204
  19. Mulesa, O. Yu., & Snityuk, V. Ye. (2020). Development of the evolutive method for forecasting hourly rows. Automation of technological and business processes, 12(3), 4–9. https://doi.org/10.15673/atbp.v12i3.1854
  20. Mulesa, O., Geche, F., Voloshchuk, V., Buchok, V., & Batyuk, A. (2017). Information technology for time series forecasting with considering fuzzy expert evaluations. In 2017 12th International Scientific and Technical Conference on Computer Sciences and Information Technologies (CSIT), 1, 105–108, IEEE, September 2017. https://doi.org/10.1109/STC-CSIT.2017.8098747
  21. Pole, A., West, M., & Harrison, J. (2018). Applied Bayesian forecasting and time series analysis. Chapman and Hall/CRC. https://doi.org/10.1201/9781315274775
  22. Taylor, S. J., & Letham, B. (2018). Forecasting at scale. The American Statistician, 72(1), 37–45. https://doi.org/10.1080/00031305.2017.1380080
  23. Tsyganok, V. V., Kadenko, S. V., & Andriichuk, O. V. (2011). Simulation of expert judgements for testing the methods of information processing in decision-making support systems. Journal of Automation and Information Sciences, 43(12). https://doi.org/10.1615/JAutomatInfScien.v43.i12.30
  24. Tsyganok, V., Kadenko, S., Andriichuk, O., & Roik, P. (2018, October). Combinatorial method for aggregation of incomplete group judgments. In 2018 IEEE First International Conference on System Analysis & Intelligent Computing (SAIC), 1–6. IEEE. https://doi.org/10.1109/SAIC.2018.8516768
  25. Wan, R., Mei, S., Wang, J., Liu, M., & Yang, F. (2019). Multivariate temporal convolutional network: A deep neural networks approach for multivariate time series forecasting. Electronics, 8(8), 876. https://doi.org/10.3390/electronics8080876
  26. Whitley, D. (1994). A genetic algorithm tutorial. Statistics and computing, 4(2), 65–85. https://doi.org/10.1007/BF00175354
  27. Wolpert, D. H., & Macready, W. G. (1997). No free lunch theorems for optimization. IEEE transactions on evolutionary computation, 1(1), 67–82. https://doi.org/10.1109/4235.585893
  28. Yan, W. (2012). Toward automatic time-series forecasting using neural networks. IEEE transactions on neural networks and learning systems, 23(7), 1028–1039. https://doi.org/10.1109/TNNLS.2012.2198074