Standard

Compression-based methods of time series forecasting. / Chirikhin, Konstantin; Ryabko, Boris.

In: Mathematics, Vol. 9, No. 3, 284, 01.02.2021, p. 1-11.

Research output: Contribution to journal › Article › peer-review

Harvard

Chirikhin, K & Ryabko, B 2021, 'Compression-based methods of time series forecasting', Mathematics, vol. 9, no. 3, 284, pp. 1-11. https://doi.org/10.3390/math9030284

APA

Chirikhin, K., & Ryabko, B. (2021). Compression-based methods of time series forecasting. Mathematics, 9(3), 1-11. [284]. https://doi.org/10.3390/math9030284

Vancouver

Chirikhin K, Ryabko B. Compression-based methods of time series forecasting. Mathematics. 2021 Feb 1;9(3):1-11. 284. doi: 10.3390/math9030284

Author

Chirikhin, Konstantin ; Ryabko, Boris. / Compression-based methods of time series forecasting. In: Mathematics. 2021 ; Vol. 9, No. 3. pp. 1-11.

BibTeX

@article{7f8309fbad6b4d5f9f3a5c8c3975de4c,
title = "Compression-based methods of time series forecasting",
abstract = "Time series forecasting is an important research topic with many practical applications. As shown earlier, the problems of lossless data compression and prediction are very similar mathemati-cally. In this article, we propose several forecasting methods based on real-world data compressors. We consider predicting univariate and multivariate data, describe how multiple data compressors can be combined into one forecasting method with automatic selection of the best algorithm for the input data. The developed forecasting techniques are not inferior to the known ones. We also propose a way to reduce the computation time of the combined method by using the so-called time-universal codes. To test the proposed techniques, we make predictions for real-world data such as sunspot numbers and some social indicators of Novosibirsk region, Russia. The results of our computations show that the described methods find non-trivial regularities in data, and time universal codes can reduce the computation time without losing accuracy.",
keywords = "Artificial intelligence, Data compression, Time series forecasting, Universal coding",
author = "Konstantin Chirikhin and Boris Ryabko",
note = "Funding Information: Funding: This research was funded by RFBR, project numbers 19-37-90009, 19-47-540001. Publisher Copyright: {\textcopyright} 2021 by the authors. Li-censee MDPI, Basel, Switzerland. Copyright: Copyright 2021 Elsevier B.V., All rights reserved.",
year = "2021",
month = feb,
day = "1",
doi = "10.3390/math9030284",
language = "English",
volume = "9",
pages = "1--11",
journal = "Mathematics",
issn = "2227-7390",
publisher = "MDPI AG",
number = "3",

}

RIS

TY - JOUR

T1 - Compression-based methods of time series forecasting

AU - Chirikhin, Konstantin

AU - Ryabko, Boris

N1 - Funding Information: Funding: This research was funded by RFBR, project numbers 19-37-90009, 19-47-540001. Publisher Copyright: © 2021 by the authors. Li-censee MDPI, Basel, Switzerland. Copyright: Copyright 2021 Elsevier B.V., All rights reserved.

PY - 2021/2/1

Y1 - 2021/2/1

N2 - Time series forecasting is an important research topic with many practical applications. As shown earlier, the problems of lossless data compression and prediction are very similar mathemati-cally. In this article, we propose several forecasting methods based on real-world data compressors. We consider predicting univariate and multivariate data, describe how multiple data compressors can be combined into one forecasting method with automatic selection of the best algorithm for the input data. The developed forecasting techniques are not inferior to the known ones. We also propose a way to reduce the computation time of the combined method by using the so-called time-universal codes. To test the proposed techniques, we make predictions for real-world data such as sunspot numbers and some social indicators of Novosibirsk region, Russia. The results of our computations show that the described methods find non-trivial regularities in data, and time universal codes can reduce the computation time without losing accuracy.

AB - Time series forecasting is an important research topic with many practical applications. As shown earlier, the problems of lossless data compression and prediction are very similar mathemati-cally. In this article, we propose several forecasting methods based on real-world data compressors. We consider predicting univariate and multivariate data, describe how multiple data compressors can be combined into one forecasting method with automatic selection of the best algorithm for the input data. The developed forecasting techniques are not inferior to the known ones. We also propose a way to reduce the computation time of the combined method by using the so-called time-universal codes. To test the proposed techniques, we make predictions for real-world data such as sunspot numbers and some social indicators of Novosibirsk region, Russia. The results of our computations show that the described methods find non-trivial regularities in data, and time universal codes can reduce the computation time without losing accuracy.

KW - Artificial intelligence

KW - Data compression

KW - Time series forecasting

KW - Universal coding

UR - http://www.scopus.com/inward/record.url?scp=85100494878&partnerID=8YFLogxK

U2 - 10.3390/math9030284

DO - 10.3390/math9030284

M3 - Article

AN - SCOPUS:85100494878

VL - 9

SP - 1

EP - 11

JO - Mathematics

JF - Mathematics

SN - 2227-7390

IS - 3

M1 - 284

ER -

ID: 27734943