Uneven spatial innovation development of Russia is due to many factors such as GRP volume, fiscal capacity of territories, fixed capital investments attracted by enterprises. However, the key factors determining the concentration of innovation industries in territorial systems of various levels are enterprises’ expenditure on innovation activity and the available scientific personnel potential. The increasing spatial heterogeneity of localization and concentration of these resources, according to our research hypothesis, enhances the spatial heterogeneity of innovation development in Russia. To confirm this hypothesis, we aim to assess the spatial heterogeneity of enterprises’ innovation development at the national level and carry out scenario modeling and forecasting of the dynamics of this heterogeneity until 2025. The paper presents a methodological approach to scenario forecasting of the spatial heterogeneity of innovation development of Russia. In the framework of the approach, the heterogeneity is assessed using spatial autocorrelation analysis according to P. Moran’s method, regression analysis of the dependence of the volume of shipped innovation goods and services performed on the costs of innovation activities carried out by enterprises, and the number of research personnel in the regions, as well as autoregressive analysis of the dynamics of their changes using a moving average (ARIMA modeling) to form the most likely forecast scenarios of innovation development for different groups of regions. The novelty of the approach lies in the system-wide use of spatial autocorrelation analysis methods based on various spatial weight matrices, regression analysis methods based on panel data and ARIMA modeling, which in combination with each other make it possible to determine the degree of influence of the factors on the heterogeneity of innovation development in regions and to form a system of various forecast scenarios. The results of the study will serve as the basis for the formation of Russia’s innovation framework. The constructed forecast scenarios will help to form strategies for innovation development in Russian regions, taking into account the identified features of the spatial localization of factors that have a significant impact on innovation development
Keywords
innovation development, spatial autocorrelation, Russia’s regions, ARIMA modeling, spatial heterogeneity, expenditure on innovation activity, Cobb – Douglas function, regression modeling