Akademik Məlumat İdarəetmə Sistemi
SOCAR Proceedings, no.4, pp.130-138, 2024 (ESCI, Scopus)
This study (unlike most of the famous works) is devoted to clarifying only the capitalization factors of oil companies of a non-speculative nature. Such factors are directly related to the main key production and financial performance indicators of companies. Today, PJSC NK Rosneft is the largest Russian oil company in terms of proven hydrocarbon reserves, occupying a leading position in the world. However, in the analyzed period, the company experienced a high degree of variability in the return on equity (6.9% in 2020 and 24.2 / 23.6% in 2021 / 2023). Therefore, at the current stage of development, monitoring the return on equity based on high-precision methods of economic and statistical modeling is relevant for PJSC NK Rosneft. Within the framework of this study, a hypothesis is put forward about the possibility of an adequate approximation and forecasting of the return on equity of a company using the Dupont model using artificial intelligence. Empirically, according to the data of the RAS of PJSC Rosneft for 2011-2023, this hypothesis is confirmed. The formed Bayesian ensemble of 7 neuromodels of various configurations allows not only to deepen the factor analysis of the company's return on equity, but also to predict its value with a high degree of accuracy (the maximum approximation error of any multilayer perceptron does not exceed 0.6%). The results obtained can be taken into account by the management of PJSC Rosneft in the process of planning the profitability of its own funds.