Methods of Mathematical Statistics in Reconstruction of Historical Data on Economic Growth Factors of the Former USSR Republics

Abstract

The lack of a complete set of historical statistical data is one of the main problems that does not allow econometric modelling and forecasting to be carried out in full. The historical data are characterized by numerous omissions, either singular or almost entire decades. The task of reconstructing them adequately is therefore always relevant. The purpose of the article is to demonstrate the application of mathematical statistics methods for the reconstruction of the series of economic indicators of the former Soviet republics, subdivided in three groups: the volume of the economy, its institutional environment and the general technological level. The authors set and solved the following tasks: 1) formed a set of data on the economic statistics of the Union republics of the former USSR; 2) demonstrated the possibility of applying mathematical methods, especially regression analysis, for the reconstruction of historical and economic statistics; 3) tested the results by determining which trends in spatial differentiation prevailed in the late USSR: towards convergence or towards divergence of the republics. Methods. Modelling was based on historical national accounts, population and labour force, human capital, science financing in the Soviet republics, wage differentials for knowledge workers and industrial workers, homicide rate, and infant mortality rate. Methods of system analysis and econometric modelling were used, mainly correlation and regression analysis. Results. Data series were reconstructed for the former Soviet republics: GDP, homicide rate, wage differentials for knowledge workers and industrial workers, infant mortality rate. Their values are given in the Appendix to the article for further use by researchers.