Subject. This article deals with the issues related to improving the methods of assessing market risk over long time horizons. Objectives. The article aims to develop a methodology for assessing VaR that takes into account dependencies in the data and adjusts the standard assumptions underlying the traditional approach to VaR calculation. Results. The article presents a constructed model for assessing long-term VaR, utilizing time series and machine learning methods including Prophet, CatBoost, VAR, and VECM. The author-developed approach offers an alternative to standard VaR methods for long-term risk assessment, allowing for the consideration of complex relationships in the data. Conclusions. The application of machine learning improves the accuracy of forecasts and enables more reliable prediction of market risks. This is especially important for long-term investors, such as pension funds and institutional investors, who need tools to manage risks in times of high volatility.
Keywords: Value at Risk, machine learning, time series, autocorrelation, stress testing
References:
Serova A.A. [The concept of Value at Risk]. Ekonomika i sotsium, 2017, no. 5, pp. 128–130. (In Russ.) EDN: ZEMGLL
Silova E.V. [Modeling of market risk]. Vestnik Bashkirskogo universiteta = Bulletin of Bashkir University, 2017, vol. 22, no. 4, pp. 925–929. URL: Link (In Russ.)
Buvaev B.L. [VaR – as a tool for assessing financial risks]. Innovatsii i investitsii = Innovations and Investments, 2018, no. 9, pp. 292–294. URL: Link (In Russ.)
Orlova L.N., Sayakhetdinov A.R. [Methods of quantitative risk assessment based on VaR: comparative analysis]. Innovatsii. Investitsii =Intellect. Innovation. Investment, 2023, no. 2, pp. 63–74. (In Russ.) DOI: 10.25198/2077-7175-2023-2-63 EDN: LGSRSH
Malkina M.Yu., Yakovleva E.K. [An analysis of growth drivers of prices for the Russian companies' shares]. Regional'naya ekonomika: teoriya i praktika = Regional Economics: Theory and Practice, 2019, vol. 17, no. 1, pp. 183–200. (In Russ.) DOI: 10.24891/re.17.1.183 EDN: YTGVID
Evdokimov M.A. [The place of key information documents in the legal regulation of financial markets in Russia and the EU]. Vestnik ekonomicheskoi bezopasnosti = Bulletin of Economic Security, 2023, no. 2, pp. 52–58. (In Russ.) DOI: 10.24412/2414-3995-2023-2-52-58 EDN: FZVDPE
Kozlova S.Yu. [Issues of public disclosure of information on structured debt financial instruments]. Teoriya i praktika obshchestvennogo razvitiya = Theory and Practice of Social Development, 2018, no. 2, pp. 34–38. (In Russ.) DOI: 10.24158/tipor.2018.2.6 EDN: YOCKPQ
Aliaskarova Zh.A., Asadulaev A.B., Pashkus V.Yu. [Forecasting the dynamics of investments in fixed assets and gross value added based on the VAR and VECM models]. Problemy sovremennoi ekonomiki = Problems of Modern Economics, 2020, no. 4, pp. 41–45. (In Russ.) EDN: UWMTAT
Shlychkov V.V., Nestulaeva D.R., Zareznov D.A. [The Russian economy in 2022: challenges and mechanisms for overcoming them]. Vestnik Chelyabinskogo gosudarstvennogo universiteta = Bulletin of Chelyabinsk State University, 2023, no. 3, pp. 240–249. (In Russ.) EDN: XXQTKM
