| Summary: | Nowadays, digital technologies influence all sectors of the economy with increasing level of introducing of modern methods of data analysis and mathematical modeling in economics and business. Complex mathematical models are developed to predict socio-economic indicators and implemented in specialized information systems.
The article describes a system of hybrid models ‘SGM Horizon’ as intellectual forecasting information system. The system of forecasting models includes a set of regression models and an expandable set of intelligent models, including artificial neural networks, decision trees, etc. Regression models include systems of regression equations that describe the behavior of forecast indicators of the development of the Russian economy in the system of national accounts. The functioning of the system of equations is determined by scenario conditions set by expert. For those indicators whose forecasts do not meet the requirements of quality and accuracy, intelligent models based on machine learning are used.
Using the ‘SGM Horizon’ tools, predictive calculations were performed for a system of 150 indicators of the social sphere of the Russian Federation using hybrid models, and for 20 indicators a significant increase in the quality and accuracy of the forecast was achieved with artificial neural network models and models of regression decision trees.
The process of models building requires considerable time: the expert must conduct numerous machine experiments with various types of models and set manually configuration parameters. In this regard, the authors see the further development of the system in the application of the multi-criteria ranking method for fuzzy objects when choosing a model, which will allow using alternative criteria to select alternative forecasting models. A scheme for the functioning of the developed system is proposed.
The forecasting system ‘SGM Horizon’ is used in the educational process of training masters in their projecting and research work.
|
|---|
