Year: 2024
Type: article
Abstract: The article is dedicated to the study of systematic risk using the Russian stock market as an example. It proposes assessing risk based on intertemporal dynamic beta using modern multivariate models such as DCC-GARCH, GJR-DCC-GARCH, and ADCC-GARCH. The analysis relies on daily returns of the MSCI World index, the main index, and eight sectoral indices of the Moscow Exchange from December 13, 2019, to June 5, 2024. The primary data consisted of daily return time series of the MSCI World Index, the main index, and eight sectoral indices of the Moscow Exchange over the period from December 13, 2019, to June 5, 2024. GARCH model construction for the return series revealed heteroscedasticity, stationarity, and deviations from a normal distribution. Based on a comparison of models with normal and Student's t-distribution in terms of predictive accuracy using cross-validation and the Diebold-Mariano test, the ADCC-GARCH model with a normal distribution was identified as providing the most accurate forecast. Through comparison of the Akaike information criterion and log-likelihood among models, GJR-DCC-GARCH was determined to be the most accurate. Analysis of the ADCC-GARCH model indicated that transport and financial indices are most susceptible to recent shocks and negative news, while the telecommunications sector is the least sensitive. The ADCC-GARCH model found that the electricity sector index has the highest conditional correlation sensitivity to the global market, while the transport index exhibits long-term correlation “memory.” Based on the descriptive statistics of Moscow Exchange beta indices and the Jarque-Bera test, the oil and gas index showed the least extreme fluctuations. Visual analysis of MSCI World and Moscow Exchange index returns revealed a declining trend in correlation between the global and Russian stock markets, indicating a deglobalization of the Russian economy.
Authors: S. P. Syrygin, E. A. Volokhin
Institution: Izhevsk State Technical University
Language: English
Subfield: Finance
Domain: Social Sciences
Open Access status: diamond
OpenAlex ID: W4405890942