Technique for Formal Methods of Time Series Forecasting Integration in Data Assimilation

Yulia Timoshenkova, Sergey Porshnev, Nikolai Safiullin

Abstract


The article describes the method developed by the authors for integration of formal methods of time series (TS) forecasting (autoregressive integrated moving average (ARIMA), singular spectrum analysis, group method of data handling, artificial recurrent neural network with long shortterm memory) into the Data Assimilation (DA), in cases where mathematical model of the dynamic system generating the TS is not known (for example TS consisting of economic indicators). The performance of the proposed integration method is illustrated on the example of forecasting TS named "Air Passengers" using the DA method based on the ensemble Kalman filter with integrated ARIMA method. Estimations of the accuracy of "Air Passengers" TS forecasts is calculated using ARIMA and the proposed integration method. Article discusses the advantages and disadvantages of the proposed method for integrating formal TS forecasting methods into the DA and also directions for its further improvement.

Full Text:

PDF (Russian)

References


G. E. Box, G. M. Jenkins, G. C. Reinsel, i G. M. Ljung, Time series analysis: forecasting and control. John Wiley & Sons, 2015.

L. Nerger i dr., «SANGOMA: Stochastic Assimilation for the Next Generation Ocean Model Applications EU FP7 SPACE-2011-1 project 283580».

M. Kanamitsu, «Description of the NMC global data assimilation and forecast system», Weather and forecasting, t. 4, vyp. 3, ss. 335–342, 1989.

A. V. Kostrov i A. M. Mardjuk, «Ob odnoj matematicheskoj zadache teorii dvizhenija raket», Zhurnal vychislitel'noj matematiki i matematicheskoj fiziki, t. 34, vyp. 10, ss. 1427–1443, 1994.

L. Kilian i M. P. Taylor, «Why is it so difficult to beat the random walk forecast of exchange rates?», Journal of International Economics, t. 60, vyp. 1, ss. 85–107, 2003.

G. Welch i G. Bishop, «An introduction to the Kalman filter», 1995.

G. P. Zhang, «Time series forecasting using a hybrid ARIMA and neural network model», Neurocomputing, t. 50, ss. 159–175, 2003.

N. Golyandina, V. Nekrutkin, i A. A. Zhigljavsky, Analysis of time series structure: SSA and related techniques. CRC press, 2001.

A. G. Ivahnenko, Dolgosrochnoe prognozirovanie i upravlenie slozhnymi sistemami. Tehnika, 1975.

Y. Timoshenkova, S. Porshnev i N. Safiullin, «On the possibility of correction of the forecasting of the Lorenz attractor dynamic characteristics using experimental data and data assimilation», v Journal of Physics: Conference Series, 2018, t. 1053, s. 012004.

Y. Timoshenkova, S. Porshnev, N. Safiullin i O. Ponomareva, «On the Possibility of the Forecast Correction Lorenz Systeme Use Particle Filter», v 2018 International Conference on Applied Mathematics and Computational Science, ICAMCS. NET 2018, 2018, ss. 27–30.

P. L. Houtekamer i H. L. Mitchell, «Data assimilation using an ensemble Kalman filter technique», Monthly Weather Review, t. 126, vyp. 3, ss. 796–811, 1998.

M. Ghil i P. Malanotte-Rizzoli, «Data assimilation in meteorology and oceanography», Advances in geophysics, t. 33, ss. 141–266, 1991.

B. K. W. Lahoz i R. Menard, Data assimilation. Springer, 2010.

E. N. Lorenz, «Deterministic nonperiodic flow», Journal of atmospheric sciences, t. 20, vyp. 2, ss. 130–141, 1963.

M. Tabor, Haos i integriruemost' v nelinejnoj dinamike. 2001.

M. Roth, G. Hendeby, C. Fritsche, i F. Gustafsson, «The Ensemble Kalman filter: a signal processing perspective», EURASIP Journal on Advances in Signal Processing, t. 2017, vyp. 1, ss. 1–16, 2017.

«Air Passengers». https://kaggle.com/abhishekmamidi/air-passengers (prosmotreno dek. 01, 2021).

M. Bocquet, «Introduction to the principles and methods of data assimilation in geosciences», Notes de cours, École des Ponts ParisTech, 2014.

Glavnye komponenty vremennyh rjadov: metod «Gusenica» / Pod red. D.L.Danilova, A.A.Zhigljavskogo. 1997. Prosmotreno: dek. 20, 2021. [Onlajn]. Dostupno na: https://rusneb.ru/catalog/001980_000024_RU_FESSL_MAIN_1331818796268482813/

A. N. Tyrsin, «Postroenie modelej avtoregressii vremennyh rjadov pri nalichii pomeh», Matematicheskoe modelirovanie, t. 17, vyp. 5, ss. 10–16, 2005.

A. E. Permanasari, I. Hidayah, i I. A. Bustoni, «SARIMA (Seasonal ARIMA) implementation on time series to forecast the number of Malaria incidence», v 2013 International Conference on Information Technology and Electrical Engineering (ICITEE), 2013, ss. 203–207


Refbacks

  • There are currently no refbacks.


Abava  Кибербезопасность IT Congress 2024

ISSN: 2307-8162