DOI: https://doi.org/10.31319/2519-8106.2(41)2019.185017

THE SOFTWARE DEVELOPMENT FOR TIME SERIES FORECASTING WITH USING ADAPTIVE METHODS AND ANALYSIS OF THEIR EFFICIENCY

Анастасія Олегівна Долгіх, Олег Григорович Байбуз

Анотація


Searching algorithm for optimal values of the smoothing coefficients of adaptive models of time series forecasting by the genetic algorithm is described. The results of the proposed approach in forecasting financial indicators are presented. The analysis of the effectiveness results of the developed algorithm with the help of a multi-criterion procedure is carried out, which allows to consider the accuracy of forecasts, the complexity of the model and to conduct analysis of adequacy using Fisher test, the determination coefficient and the mechanism for checking the residues.

Ключові слова


forecasting; adaptive methods; genetic algorithm; analysis of model quality

Повний текст:

PDF

Посилання


Chuchuieva, I.A. (2012). Model' prohnozuvannya chasovykh ryadiv za vybirkoyu maksymal'noyi pravdopodoby: dysertatsiya na otrymannya stupenyu kandydata tekhnichnykh nauk [A time series forecasting model for a maximum likelihood sample: thesis for a candidate of technical sciences]. Candidate’s thesis. Moscow: Moscow State Technical University named after Bauman.

Tikhonov, E.E. (2006). Metody prohnozuvannya v umovakh rynku [Forecasting methods in market conditions]. Nevinnomyssk.

Yang, J.M. (2006). Power System Short-term Load Forecasting. Candidate’s thesis. Germany, Darmstadt: Elektrotechnik und Informationstechnik der Technischen Universitat.

Biloborodko, O.I., T.G. Yemelyanenko. (2014). Analiz dynamichnykh ryadiv [Dynamic Series Analysis]. Dnipro: DNU named after Oles Holnchar [in Ukrainian].

Osman Dag, Ceylan Yozgatligil. (2016). GMDH: An R Package for Short Term Forecast-ing via GMDH-Type Neural Network Algorithms. The R Journal Vol. 8/1, Aug. 2016, ISSN 2073-4859. Retrieved from https://journal.r-project.org/archive/2016/RJ-2016-028/RJ-2016-028.pdf.

Dolhikh A.O. (2019). Prohnozuvannya finansovykh chasovykh ryadiv z vykorystannyam neyronnoyi merezhi typu MHUA [Financial time series forecasting using a neural network of the GMDH type]. Proceedings from XXIII mizhnarodnoho molodizhnoho forumu «Radiolektreronika i molodʹ u XXI stolitti», konferentsiya «Matematychne modelyuvannya, systemnyy analiz i teoriya optymal’nykh rishen’» – The Twenty Third International Youth Forum “Radio Electronics and Youth in the XXI Century”, Conference “Mathematical Modeling, Systems Analysis and Theory of Optimal Solutions”. (pp. 146–147). Kharkiv [in Ukranian].

The group method of data handling (GMDH). Retrieved from http://www.mgua.irtc.org.ua/ru/index.php?page=gmdh (application date: 09.05.2019).

A.V.Tregub, I.V.Tregub. (2011). Metodyka pobudovy modeli ARIMA dlya prohnozuvannya dynamiky chasovykh ryadiv. [The method of constructing the ARIMA model for predicting the dynamics of time series]. «Lisovyy visnyk», T.5, rozdil «Matematychne modelyuvannya» – “Forecast Newspaper”, 5, section “Mathematical modeling”. (pp. 179–183).

Hyndman, R.J., Khandakar, Y. (2008). Automatic Time Series Forecasting: The forecast Package for R. Journal of Statistical Software. July 2008. Vol. 27, Issue 3.

Lukashin, Yu.P. (2003). Adaptyvni metody korotkostrokovoho prohnozuvannya chasovykh ryadiv [Adaptive methods of time series short-term forecasting]. M.: Finance and Statistics.

Herrera F., Lozano M., Verdegay J.L. (1998). Tackling real-coded Genetic algorithms: operators and tools for the behavior analysis. Artificial Intelligence Review, Vol. 12, No. 4, (pp. 265–319).

YAHOO! Finance. Retrieved from https://finance.yahoo.com/.

Kwiatkowski, D., Phillips P.C.B., Schmidt, P., Shin, Y. (1992). Testing the null hypothesis of stationarity against the alternative of a unit root. Journal of Econometrics, 54. (pp. 159–178).

Armstrong, J.S., Adya, M., Collopy, F. (2001). Rule-Based Forecasting Using Judgment in Time Series Extrapolations. Principles of Forecasting: A Handbook for Researcher and Practitioners. Kluwer, Dordrecht.

Wichard, J.D., Ogorzalek, M. (2007). Time series prediction with ensemble models. Neurocomputing, 70, Issues 13–15. (pp. 2371–2378).




Текст

ISSN 2519-8106 (Print), eISSN 2519-8114 (Online)