Informatics and Applications

2022, Volume 16, Issue 2, pp 11-18

MODEL SETTING USING STATIONARITY CRITERIA FOR TIME SERIES FORECASTING

O. A. Kravtsova

Abstract

The article discusses the possibility of using the information on the stationarity of residuals to improve the procedure of forecasting nonstationary time series. In the traditional approach, this procedure is used only to confirm or reject the hypothesis of nonstationarity of residuals. In this article, the stationarity test is used for fine-tuning of hyperparameters of the forecasting models. The technique is based on the Granger cointegration approach property to find a statistically significant relationship between time series. The author used the p-value of stationarity tests as a loss function. Economic and generated time series were used as data for verification. The experiments have shown that this approach is often more effective in comparison with the traditional methods of tuning models.

[+] References (23)

Khmelnitskaya, A. B. 2002. Social welfare functions for different subgroup utility scales. Constructing and applying objective functions. Eds. A. Tangian, and J. Gruber. Lecture notes in economics and mathematical systems ser. Berlin, Heidelberg: Springer-Verlag. 510:515-530. doi: 10.1007/978-3-642-56038-5.
Engle, R. F, and C.W.J. Granger. 1987. Co-integration and error correction: Representation, estimation, and testing. Econometrica 55(2):251-276. doi: 10.2307/ 1913236.
Hatemi-J, A. 2008. Tests for cointegration with two un-known regime shifts with an application to financial market integration. Empir. Econ. 35:497-505. doi: 10.1007/ s00181-007-0175-9.
Ariyo, A. A., A. O. Adewumi, and C. K. Ayo. 2014. Stock price prediction using the ARIMA model. 16th Conference (International) on Computer Modeling and Simulation Proceedings. Piscataway, NJ: IEEE. 106-112. doi: 10.1109/UKSim.2014.67.
Li, T, J. Zhong, and Z. Huang. 2019. Potential dependence of financial cycles between emerging and developed countries: Based on ARIMA-GARCH copula model. Emerg. Mark. Financ. Tr. 56(6):1-14. doi: 10.1080/ 1540496X.2019.1611559.
Gepp, A., K. Kumar, and S. Bhattacharya. 2010. Business failure prediction using decision trees. J. Forecasting 29(6):536-555. doi: 10.1002/for.1153.
Aggarwal, C. C. 2015. Mining time series data. Data min-ing. Springer. 457-491.
Namini, S. S., N. Tavakoli, andA. S. Namin. 2018. Acom- parison of ARIMA and LSTM in forecasting time series. 17th Conference (International) on Machine Learning and Applications Proceedings. Piscataway, NJ: IEEE. 536-555. doi: 10.1109/ICMLA.2018.00227.
Temur, A. S., M. Akgun, and G. Temur. 2019. Predicting housing sales in Turkey using ARIMA, LSTM and hybrid models. J. Bus. Econ. Manag. 20(5):920-938. doi: 10.3846/jbem.2019.10190.
Wang, Z., and Y. Lou. 2019. Hydrological time series forecast model based on wavelet de-noising and ARIMA-LSTM. 3rd Information Technology, Networking, Electronic and Automation Control Conference Proceed-ings. Piscataway, NJ: IEEE. 1697-1701. doi: 10.1109/ ITNEC.2019.8729441.
Ye, T. 2017. Stock forecasting method based on wavelet analysis and ARIMA-SVR model. 3rd Conference (International) on Information Management Proceedings. Piscataway, NJ: IEEE. 102-106. doi: 10.1109/ INFOMAN.2017.7950355.
Singh, S., K. S. Parmar, J. Kumar, and S. J. S. Makkhan. 2020. Development of new hybrid model of discrete wavelet decomposition and autoregressive integrated mov-ing average (ARIMA) models in application to one month forecast the casualties cases of COVID-19. Chaos Soliton. Fract. 135(1):109866. doi: 10.1016/j.chaos.2020.109866. 8 p.
Zhou, F., Q. Zhang, D. Sornette, and L. Jiang. 2019. Cascading logistic regression onto gradient boosted decision trees for forecasting and trading stock indices. Appl. SoftComput. 84(2):105747.13 p. doi: 10.1016/j.asoc. 2019.105747.
Ribeiro, M. H. D. M., andL. Coelho. 2020. Ensemble ap-proach based on bagging, boosting and stacking for short-term prediction in agribusiness time series. Appl. Soft Com - put. 86:105837. 17 p. doi: 10.1016/j.asoc.2019.105837.
Elder, A. 2014. The new trading for a living: Psychology, discipline, trading tools and systems, risk control, trade management. Wiley Trading ser. New York, NY: John Wiley & Sons Ltd. 304 p.
Tabachnick, B. G., and L. S. Fidell. 2013. Using multivari-ate statistics. 6th ed. London: Pearson Education. 983 p.
Hyndman, R., and Y. Khandakar. 2008. Automatic time series forecasting: The forecast package for R. J. Stat. Softw. 27(3):1-22. doi: 10.18637/jss.v027.i03.
Pedregosa, F., G. Varoquaux, A. Gramfort, et al. 2011. Scikit-learn: Machine learning in Python. Decision Trees. J. Mach. Learn. Res. 12:2825-2830.
Singh, S., and P. Gupta. 2014. Comparative study Id3, cart and C4.5 decision tree algorithm: A survey. Int. J. Ad-vanced Information Science Technology 3(7):47-52. doi: 10.15693/ijaist/2014.v3i7.
Dickey, D.A., and WA. Fuller. 1979. Distribution of the estimators for autoregressive time series with a unit root. J. Am. Stat. Assoc. 74(366):427-431. doi: 10.2307/ 2286348.
Zivot, E., and D. Andrews. 2002. Further evidence on the great crash, the oil-price shock, and the unit-root hy-pothesis. J. Bus. Econ. Stat. 10(3):251-270. doi: 10.2307/ 1391541.
Finam. Dannye ot 01.01.2005 do 20.11.2020 [Data from 01.01.2005 to 20.11.2020]. Available at: https://www. finam.ru/ (accessed April 6, 2022).
Litvinova, Ya. S. 2012. Tsena na neft' kak klyuchevoy faktor vozdeystviya na rossiyskuyu valyutu [Oil price as a key factor affecting the Russian currency]. Mat-ly I Mezhdunar. nauchn. konf. "Problemy i perspektivy ekonomiki i upravleniya" [I Scientific Conference (International) "Problems and Prospects of Economics and Management" Proceedings]. St. Petersburg: Renome. 19-21. Available at: https://moluch.ru/conf/econ/archive/15/ 2170 (accessed June 3, 2022).

[+] About this article