The regional differentiation makes impossible the sustainable socio-economic development of the subjects of the Russian Federation without the monitoring public governance results in space and time. Despite the comprehensive approach of the current procedure, approved by the federal government, it does not adequately assess the executive authorities effectiveness. Its main problem is the impossibility to assume such important administrative function as forecasting the social and economic development of Russian territorial subjects. The authors propose an alternative methodology on the basis of the system economic theory. This technique is implemented in several consecutive stages. Firstly, we develop the system of 30 indicators. Secondly, we normalize the values of the indicators using the method of pattern. Thirdly, we calculate the index of the social and economic development of Russian regions for 2011–2015 assuming that the indicators are equal. Last, we group Russian regions into clusters according to the level of their social and economic development using neural network technologies (Kohonen selforganizing maps). Only 9 in 80 subjects of the Russian Federation (RF) had the degree of realizing the social and economic potential higher than 40 % during the period under consideration. In 2011–2015, the most of regions had a low and lower than average level of social and economic development (with an aggregate share about 64.3 %). It means that, under current conditions, the majority of the RF regions have considerable reserves for realizing their social-economic potential. In particular, the absence of the territorial subjects with a high level of social and economic development proves that. The authors have simulated the social and economic situation of the RF subjects by means of an adequate Bayesian neural networks. The obtained results can be used as the basis for further research in the field of evaluating executive authorities effectiveness and forecasting the level of social and economic development of Russian regions.

The industry plays a key role in Russia’s economic security. Significant interregional gaps in the level of socio-economic development made topical the issue of optimizing the distribution of production forces in the constituent entities of the Russian Federation through increasing the efficiency of investment and industrial policy at the meso-level. In our opinion, the efficiency of such policy is impossible without modern methods of economic and mathematical modelling and information technologies. The Cobb-Douglas production function is still an adequate method for an accurate assessment of production capacities of the industry both in the whole country and it’s regions. At present, such method is being actively developed in two directions: the function’s modification through “saturation” with different factors and improvement of the approaches to determining its dynamic (changing in time) parameters. In this study, we hypothesized the possibility of building an adequate Cobb-Douglas production function with static and dynamic parameters, using the case of the Republic of Bashkortostan’s industry for the period from 2006 to 2016. The first hypothesis about using static parameters for building Cobb-Douglas production function was empirically rejected. In contrast, we have confirmed the second hypothesis. We defined dynamic parameters of the Cobb-Douglas production function using the alternative method of linear regression (AMLR). This method accurately assesses production capacities of the Republic of Bashkortostan. The choice of the method is not random. First of all, its use allows, in any case, ensuring a correct economic sign of parameters with factor indicators (labour and capital). Secondly, the original method of calculations, using the growth rates of the indicators, ensures high accuracy of verifying the model parameters. Thus, the conducted calculations have shown that the dynamics of the regular component explains 57.4 % of variance of the initial time series (remainder functions after defining the mode l parameter s using the AMLR) . Combine d with other econometric methods , the application of the developed model will also enable an accurate forecast of the production capacities of the industry in regions. The study’s results can be applied for developing the investment and industrial policy of the Republic of Bashkortostan.