Markov chains in the task of predicting the values of a time series
- 1 Financial University under the Government of the Russian Federation (Moscow, Russia)
Abstract
A sample of the log yield values of the shares of a Russian company is considered. The problem of estimating the average time required to achieve a given final value of log profitability, with a known initial value, is solved. To solve the problem: a Markov chain is constructed from the enlarged states of the time series; the probabilities of transition of the Markov chain are estimated from a sample; the average time of achievement is calculated from the matrix of transition probabilities. The practical implementation of the algorithm is carried out by means of the Python programming language.
- Received
- 11.08.2023
- Approved after review
- 25.08.2023
- Accepted
- 02.10.2023
How to cite
Криволапов С.Я. Марковские цепи в задачах прогноза значений временного ряда // Информационно-измерительные и управляющие системы. 2023. Т. 21. № 6. С. 48−54. DOI: https://doi.org/10.18127/j20700814-202306-06
References
- Elchenkov R.A., Dunaev M.E., Zaitsev K.S. Time series forecasting in real-time streaming data processing. International Journal of Open Information Technologies. 2022. 1. S. 63−69.
- Lobach S.V., Zhuk E.E. Prognozirovanie vremennykh ryadov markovskogo tipa na osnove veivlet-analiza. Materialy Mezhdunar. nauchnoi konf. "Informatsionnye tekhnologii i sistemy 2014" (ITS 2014). Minsk: BGUIR. 2014. S. 282−283. (in Russian)
- Lobach S.V., Zhuk E.E. Ob odnom sposobe prognozirovaniya vremennykh ryadov s propuskami. Ekonomika, modelirovanie, prognozirovanie. 2016. 10. S. 218−223. (in Russian)
- Makkini Ues. Python i analiz dannykh. M.: DMK Press. 2020. 482 s. (in Russian)
- Krivolapov S.Ya., Khripunova M.B. Matematika na Python. M.: KNORUS. 2021. 456 s. (in Russian)
- Chzhun Kai-Lai. Odnorodnye tsepi Markova. M.: Mir. 1964. 427 s. (in Russian)
- Shiryaev A.N. Veroyatnost−2. M.: MTsNMO. 2011. 968 c. (in Russian)
- Koralov L.B., Sinai Ya.G. Teoriya veroyatnostei i sluchainye protsessy. M.: MTsNMO. 2013. 408 s. (in Russian)
- Vydrin A.S., Mikhalev A.V. Stokhasticheskie matritsy i analiz zashchishchennosti avtomatizirovannykh sistem. Fundamentalnaya i prikladnaya matematika. 2007. T. 13. № 1. S. 61−99. (in Russian)
- Krivolapov S.Ya. Markovskie tsepi i yazyk Python v zadachakh dostizheniya zadannogo sostoyaniya sistemy. Myagkie izmereniya i vychisleniya. 2022. T. 61. № 12. S. 19−31. (in Russian)