Папина О. Н.

S u b j e ct . Methods of forecasting target parameters of socio - economic development of regions. R e s e a r c h topic. Forecasting the export of Russian regions using the standard apparatus of neural networks. P u r p o s e. Development and testing of a methodological approach to forecasting the target parameters of the socio - economic development of Russian regions for the period up to 2024. M e t h o d o l o g y. Logical, comparative, economic - statistical, cluster analysis, neural network prediction. R e s u l t s . Five virtual clusters (groups of regions) are formed, with a sufficient level of reliability and significance of parameters. An analysis of the export dynamics of model regions representing different groups (virtual clavesters) of Russian regions, differing in the totality of parameters of foreign economic activity, was made. A forecast of the dynamics of exports of model regions has been formed. S c o p e . Economic policy of regional authorities in the field of foreign economic activity. C o n c l u s i o n s . During the study, it was found that for most region groups, the combination of cluster analysis, correlation - regression and neural network prediction methods allows: to establish the composition of basic indicators of socio - economic development associated with the target parameter; make at least two variants of target parameter dynamics forecast; to determine the degree of difference between the forecast variants of the target parameter dynamics; determine the acceptability of using formal prediction methods (correlation - regression and neural) to predict the dynamics of the target parameter of socio - economic development of model regions.

Ecological and economic development of Russian regions. Topic . Assessment of spatial and functional features of ecological and economic activity of Russian regions.Purpose. Identification of the level of ecological and economic activity and the balance of Russian regions in terms of economic, social and environmental properties. Methodology. Logical, comparative, economic and statistical, and cluster analysis. Results . Five virtual clusters (groups of regions) were formed with a sufficient level of confidence and significance of parameters. The analysis of the ratio of economic, social and environmental parameters made it possible to assess the activity and balance of regional groups within and between the respective spheres. The correlation of ten parameters (three social, three economic, and four environmental) showed that at the beginning of the analyzed period (2005), there were two groups of the most active regions with high values of indicators of ecological and economic activity. Applicationarea . Economic and environmental policy of regional authorities. Conclusions . In General, there was no link between environmental and economic activity and balance at the level of regional groups. During the period up to 2018, there were noticeable changes in the configuration of regional groups and their environmental and economic parameters. The level of balance of active clusters decreased and increased in the clusters of medium and low activity.