Comparative analysis of the forecast for variation of the parameters of the hybrid linguistic model
DOI:
https://doi.org/10.34185/1562-9945-3-122-2019-05Keywords:
часовий ряд, модель, прогнозування, гібридна лінгвістична модель, лінгвістичне моделюванняAbstract
Using a hybrid linguistic approach to model numerical images in the form of time series using probabilistic grammars based on hidden time series and implement information technology to build sets of linguistic models and their hybrids that describe the dynamics of selected time series of processes of different nature.
In the article the results of computational experiments are considered, the quality of forecasting of time series of diverse nature at various parameters was proved.
The goal of the current research is to provide empirical evidence of the suitability of using a hybrid linguistic approach for predicting time series.
Experimental way to find the optimal parameters of the algorithm. The algorithm was applied to a variety of time series (social, medical, financial and economic), calculated the statistical accuracy of the forecast. Experiments have shown that the algorithm consistently performs the forecast of values in a range of 3-4 steps forward and forecasts the trend change by 3-5 steps.
References
Fu K. Strukturnue metodu v raspoznavanyy obrazov; [per. s anhl. Z.V. Zavalyshyna, S.V. Petrova, R.L. Sheinyna; pod red. M.A. Aizermana]. M.: Myr, 1977. 319 s.
Fu K.S. Structural Methods in Pattern Recognition. Academic Press,1969,227p.
Sokolov O.Iu. Alhebraichni modeli ta metody analizu i syntezu system keruvannia slabo formalizovanymy protsesamy : Dys... d-ra nauk: 05.13.03 [Mistse zakhystu: Natsionalnyi aerokosmichnyi universyt im. M.Ie. Zhukov-skoho “Kharkivskyi aviatsiinyi instytut”]. m. Kharkiv - 2002. - 393 s.
Senkevych Yu.Y. Teoretycheskye osnovы razrabotky avtomatyzyro-vannыkh ynstrumentalnыkh sredstv dlia telemedytsynskykh system poliar-nыkh zon: dyssertatsyia ... doktora tekhnycheskykh nauk: 05.13.01 / Senkevych Yuryi Yhorevych; [Mesto zashchytы: Sankt-Peterburhskyi ynstytut ynfor-matyky y avtomatyzatsyy RAN].- Sankt-Peterburh, 2008.- 271 s.
Drozhdyn V.V., Bakanov A.B. Hrammatyka opysanyia domena famylyi // Voprosы radyoэlektronyky. Ser.:Эlektronnaia vыchyslytelnaia tekhny-ka. 2007. Vыp.1.S.77–82.
Mottl V.V., Sulymova V.V., Tatarchuk A.Y. Avtomatycheskyi vыbor naybolee ynformatyvnыkh frahmentov v zadachakh raspoznavanyia syhnalov raznoi dlytelnosty. Tavrycheskyi vestnyk matematyky y ynformatyky – No 1, 2006, str. 109-115.
Cordón O., Herrera F. A Proposal for Improving the Accuracy of Linguistic Modeling // IEEE TRANSACTIONS ON FUZZY SYSTEMS, VOL. 8, NO. – Spane, Granada, University of Granada. 3, JUNE 2000. - P.335-344.
G.-W. Wei. Uncertain Linguistic Hybrid Geometric Mean Operator and Its Application to Group Decision Making Under Uncertain Linguistic Environment // International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems. Vol. 17, No. 2 (2009) 251–267.
Baklan Y.V. Lynhvystycheskoe modelyrovanye: osnovы, metodы, ne-kotorыe prykladnыe aspektы // Systemnыe tekhnolohyy. Rehyonalnыi mezhvuzovskyi sbornyk nauchnыkh rabot. – Vыpusk 3 (74). – Dnepropet-rovsk, 2011. – S.10 – 19.
Baklan I.V. Intervalnyi pidkhid do pobudovy linhvistychnoi modeli // Systemnыe tekhnolohyy. Rehyonalnыi mezhvuzovskyi sbornyk nauchnыkh rabot. – Vыpusk 3 (86). – Dnepropetrovsk, 2013. - S.3-8.
Baklan I.V. Linhvistychne modeliuvannia chasovykh riadiv riznoi pry-rody z fraktalnymy vlastyvostiamy // Systemnыe tekhnolohyy. Rehyona-lnыi mezhvuzovskyi sbornyk nauchnыkh rabot. – Vыpusk 3 (104). – Dnep-ropetrovsk, 2016. - S.110-118.
Hibrydnyi linhvistychnyi pidkhid do modeliuvannia chasovykh riadiv / I.V.Baklan, T.V.Shulkevych // Prykladni pytannia matematychnoho modeliu-vannia. - №2 (2018). - S.192-202.
Elektronnyi resurs http://www-eio.upc.edu/~pau/cms/rdata/datasets.html.
