Search academic texts

Information × Registration Number 0224U031809, 0123U101997 , R & D reports Title Development of stochastic models, statistical methods for analysis and optimization of systems in the medical and socio-economic spheres popup.stage_title Head Rozora Iryna V., Доктор фізико-математичних наук Registration Date 18-06-2024 Organization Taras Shevchenko National University of Kyiv popup.description2 The operation of systems in steady-state mode is investigated, optimal control problems are formulated, and explicit formulas for calculating the most important indicators of system performance are found. The strategies of sequential resource allocation in the environment represented by the model of a stochastic multi-armed bandit are considered. The complexity of these strategies, the conditions of their efficiency and losses for the case of a multi-armed bandit in a stochastic environment with a Bernoulli distribution are analyzed. The considered theoretical information was used to develop software for conducting numerical experiments to demonstrate the effectiveness of the strategies. A complex of non-invasive studies of multimorbid patients with cardiac and rheumatological profile has been developed, which consists of both well-known and innovative techniques. A new metric (CG) for assessing the consistency of a set of empirical data is proposed, which has a number of advantages over the Kendall's rank concordance coefficient and allows to assess the consistency for each element of the group. Also, an original metric for determining the consistency of magnetocardiographic maps during the ST-T interval is proposed. An analysis of existing approaches and mathematical methods for solving two-level discrete optimization problems is carried out. Their application to modeling decision-making processes in complex socio-economic, environmental and financial systems is investigated. The application of switching regression models to the problem of monitoring and predicting the number of coronavirus infections is considered. The general approach to the use of this type of models and their basic assumptions are considered, and the methodology for building a time series model of the number of daily coronavirus infections is outlined. The modeling and statistical analysis of the spread of the infection are carried out.   Product Description popup.authors Vasylyk Olha I. Hryhorieva Nataliia M. Zatula Dmytro V. Knopov Pavlo S. Hanna V. Livinska Lukovych Olha V. Pererva Antonina H. Ponomarev Vadym D. Semenova Nataliia V Usar Iryna Ya. Frankov Oleksii О. Chaikovskyi Illia A. popup.nrat_date 2024-06-18 Close