Metrics for Evaluating the Dependence of Two Time Series

Andrey Gulyaev, Svetlana Pivneva

Abstract


This article evaluates the dependence of time series through a comprehensive examination of the main methods for identifying dependencies, as well as their advantages and disadvantages. Time series are sequences of observations measured at equal intervals of time and play a crucial role in data analysis across various fields of science and technology. A measure of dependence between time series is essential for understanding the relationships and interactions between different processes. This can help improve forecasting models, optimize resource management, and enhance the accuracy of diagnostic systems. The first part of the article provides a detailed review of traditional time series analysis methods, highlighting their strengths and weaknesses. The second part of the article focuses on two advanced methods: distance correlation and mutual information. These methods are modern tools that enable researchers to detect both linear and nonlinear dependencies between time series, significantly expanding the analytical potential compared to traditional approaches based on linear correlation. This article aims to promote the integration and further development of these methods in various scientific and applied contexts. The authors' efforts are directed toward facilitating their widespread application for more accurate modeling and forecasting of temporal processes, which, in turn, can significantly improve the quality and precision of analytical research.


Full Text:

PDF (Russian)

References


Zverovshchikova, N. V. Analysis of Time Series and Its Application in Economics and Finance / N. V. Zverovshchikova, I. M. Moyko, V. S. Shatov // Analytical and Numerical Methods for Modeling Natural Science and Social Problems (ACHM-2023): Collection of Articles Based on Materials from the XVIII All-Russian Scientific and Technical Conference with International Participation, Dedicated to the 80th Anniversary of Penza State University and the 80th Anniversary of the "Higher and Applied Mathematics" Department, Penza, November 6–10, 2023. – Penza: Penza State University, 2023. – P. 165-172. – EDN SAIIHF.

Muklaev, S. M. A Review of Methods for Identifying Mutual Dependencies in Time Series / S. M. Muklaev, A. S. Filippova // Interuniversity Scientific and Technical Conference for Students, Postgraduates, and Young Specialists Named After E. V. Armensky: Conference Materials, Moscow, April 5–13, 2022 / Moscow Institute of Electronics and Mathematics, National Research University "Higher School of Economics". – Moscow: Moscow Institute of Electronics and Mathematics, NRU HSE, 2022. – P. 168-170. – EDN DMOARW.

Kolosov, M. V. Development of Models for Forecasting Building Heat Consumption / M. V. Kolosov, Yu. L. Lipovka, E. E. Shishkova // Energy Security and Energy Saving. – 2023. – No. 3. – P. 17-22. – EDN JFMBNL.

Gulyaev, A. V. Determining Key Factors Influencing Consumer Demand Considered in Sales Forecasting / A. V. Gulyaev // Development of Modern Science and Technology in the Context of Transformational Processes: Collection of Materials from the IX International Scientific and Practical Conference, Moscow, February 22, 2023. – St. Petersburg: Printing Workshop, 2023. – P. 374-382. – EDN BQNGPR.

Apalkov, F. S. Searching for Dependency Trends Between Related Time Series Based on Angular Trend Clustering / F. S. Apalkov, V. P. Stepanov // Artificial Intelligence in Automated Control Systems and Data Processing: Collection of Articles from the II All-Russian Scientific Conference, in 5 volumes, Moscow, April 27–28, 2023. – Moscow: KDU, Dobrosvet, 2023. – P. 50-54. – EDN MUGCTL.

Kunin, V. A. Quantitative Assessment of the Influence of a Debtor's Financial Position on the Value of Accounts Receivable / V. A. Kunin, S. G. Konkova // Financial Business. – 2023. – No. 12(246). – P. 337-343. – EDN MVLUGS.

Petrenko, I. V. A Popular Overview of Pearson's Linear Correlation Coefficient / I. V. Petrenko // Ways to Improve the Efficiency of Public Administration in the Context of Socio-Economic Development of Territories: Materials from the VII International Scientific and Practical Conference, Donetsk, June 6–7, 2023. – Donetsk: Donetsk Academy of Management and Public Administration, 2023. – P. 63-66. – EDN DAPREY.

Bychaev, A. G. Comparative Analysis of the Use of Correlation Coefficients / A. G. Bychaev // Scientific Support for the Development of the Agro-Industrial Complex in the Context of Import Substitution: Collection of Scientific Papers from the International Scientific and Practical Conference, St. Petersburg - Pushkin, May 25–27, 2022. – St. Petersburg: St. Petersburg State Agrarian University, 2022. – P. 89-93. – EDN CACARO.

Makarenko, E. A. Application of Pearson's Linear Correlation for Forecasting the Bankruptcy of Insurance Companies / E. A. Makarenko // Current Problems of Economics and Management. – 2021. – No. 1(29). – P. 41-44. – EDN BZTSRP.

Analysis of global temperature influencing factors based on spearman correlation coefficient method and grey correlation theory / Yu. Zheng, Y. Meng, E. Lou [et al.] // Highlights in Science, Engineering and Technology. – 2023. – Vol. 48. – P. 102-111. – DOI 10.54097/hset.v48i.8271. – EDN BHEQPU.

Pavlova, K. V. Application of Spearman's Rank Correlation Coefficient in Statistical Research / K. V. Pavlova // UIS Science. – 2023. – Vol. 3, No. 3(3). – P. 28-32. – EDN UBHMWG.

Wang, H. Spearman Rank Correlation Screening for Ultrahigh-Dimensional Censored Data / H. Wang, J. Yan, X. Yan // Proceedings of the AAAI Conference on Artificial Intelligence. – 2023. – Vol. 37, No. 8. – P. 10104-10112. – DOI 10.1609/aaai.v37i8.26204. – EDN EOOPAH.

Li, W. Distance correlation test for high-dimensional independence / W. Li, Q. Wang, J. Yao // Bernoulli. – 2024. – Vol. 30, No. 4. – DOI 10.3150/23-bej1710. – EDN TGCOFI.

Sudrajat, K. B. Correlation between Running Distance and Iliotibial Band Syndrome among Yogyakarta Runners / K. B. Sudrajat, F. Rahman // Jurnal Pendidikan Jasmani dan Olahraga. – 2023. – Vol. 8, No. 1. – P. 58-66. – DOI 10.17509/jpjo.v8i1.55438. – EDN PWHNOC.

Székely, G. J. Brownian distance covariance / G. J. Székely, M. L. Rizzo // Annals of Applied Statistics. – 2009. – Vol. 3, No. 4. – P. 1236-1265. – DOI 10.1214/09-AOAS312. – EDN RLDMUV.

Das, R. Feature selection with distance correlation / R. Das, G. Kasieczka, D. Shih // Physical Review D. – 2024. – Vol. 109, No. 5. – P. 054009. – DOI 10.1103/physrevd.109.054009. – EDN BDFVYP.

Zhang, B. Distance correlation entropy and ordinal distance complexity measure: efficient tools for complex systems / B. Zhang, P. Shang // Nonlinear Dynamics. – 2024. – Vol. 112, No. 2. – P. 1153-1172. – DOI 10.1007/s11071-023-09080-8. – EDN OUPAMS.

Zaiats, V. M. Claude shannon – the founder of classical information theory and his influence on modernity / V. M. Zaiats // Computer Technologies of Printing. – 2023. – Vol. 2, No. 50. – P. 41-58. – DOI 10.32403/2411-9210-2023-2-50-41-58. – EDN ITYQFC.

Harré, M. S. Entropy, Economics, and Criticality / M. S. Harré // Entropy. – 2022. – Vol. 24, No. 2. – P. 210. – DOI 10.3390/e24020210. – EDN ZKYLAP.

Ellerman, D. Introduction to logical entropy and its relationship to Shannon entropy / D. Ellerman // 4open. – 2022. – Vol. 5. – P. 1. – DOI 10.1051/fopen/2021004. – EDN MPJRYB.

Anh, C. T. Mutual Information Based on Multiple Level Discretization Network Inference from Time Series Gene Expression Profiles / C. T. Anh, Yu. K. Kwon // Applied Sciences (Switzerland). – 2023. – Vol. 13, No. 21. – P. 11902. – DOI 10.3390/app132111902. – EDN IJLFWN.

Lapteva A.V. Information Theory / Ural State University of Economics / Yekaterinburg / 2023 / 98 pages. EDN FDMZQE.

Contreras-Reyes, J. E. Mutual information matrix based on asymmetric Shannon entropy for nonlinear interactions of time series / J. E. Contreras-Reyes // Nonlinear Dynamics. – 2021. – DOI 10.1007/s11071-021-06498-w. – EDN DBONLZ.

Multiscale permutation mutual information quantify the information interaction for traffic time series / Y. Yin, X. Wang, Q. Li [et al.] // Nonlinear Dynamics. – 2020. – Vol. 102, No. 3. – P. 1909-1923. – DOI 10.1007/s11071-020-05981-0. – EDN FCXPUW.

Yamashita Rios De Sousa, A. M. Sign patterns symbolization and its use in improved dependence test for complex network inference / A. M. Yamashita Rios De Sousa, Ja. Hlinka // Chaos (Woodbury, N.Y.). – 2023. – Vol. 33, No. 8. – DOI 10.1063/5.0160868. – EDN IXBSCB.


Refbacks

  • There are currently no refbacks.


Abava  Кибербезопасность ИБП для ЦОД СНЭ

ISSN: 2307-8162