Modelling the Dynamics of the World Stock Indices

• Main Author: Katunina Olha S.
• Format: Article
• Language: English
• Published: Research Centre of Industrial Problems of Development of NAS of Ukraine 2017-11-01
• Series: Bìznes Inform
• Subjects: stock indices; time series; dynamic factor analysis; forecasting
• Online Access: http://www.business-inform.net/export_pdf/business-inform-2017-11_0-pages-197_202.pdf

The article considers the issues of modelling the dynamics of the economic system, which is described by the aggregate of indicators from some major world stock indices. The practical possibility of application of methodology of dynamic factor analysis for researching such systems has been substantiated. The mathematical model, combining approaches of factor analysis and autoregressive estimation, has been developed. In contrast to the classical models of the vector time series analysis, a dynamic factor system is used to simulate evolution, which is more informative than the original aggregate of indicators. The basic calculated correlations of the constructed mathematical model and the direction of adaptation of the developed algorithm for solving a wide spectrum of forecasting tasks have been provided. Unlike well-known models, the proposed technique allows to define a certain interval of reasonable forecast values, the boundaries of which depend on the parameters of the model. The error in the control time interval, determined when comparing the averaged values of the forecasts for several indicators with actual index values, persists in the range of 1 – 2%, which confirms the high efficiency of the proposed methodology and the possibility of its use in research practice.