Social policy just like any other type of policy is an element of the management system. For effective social policy it is necessary to know the trends and quantitative characteristics of social development dynamics. The purpose for the research is to build an objective composite index to measure and compare the quality of life in constituent entities of the Russian Federation. Such an index is particularly useful in Russia which is undergoing the process of transformations amid increasing social and economic inequality. Objective complex assessment of the quality of life can play a major role in reducing such inequality in Russia. The author implements the algorithm of building a latent integral characteristic of system quality change based on statistical indicators for a series of sequential observations on the basis of principal components analysis given the measurement noise (SNR-based algorithm). As opposed to the classic principal components analysis where the information capacity of the calculated integral characteristic set a priori and is provided by selecting a number of principal components, in the proposed algorithm the information capacity is estimated a posteriori, based on variance criteria and the selected signal-to-noise ratio characterizing data variability. When building an objective indicator of the quality of life the author uses methods of multivariate analysis and the signal detection theory. With the use of the proposed algorithm the author builds integral indicators of the quality of life of constituent entities of the Russian Federation in 2007–2014. It is concluded that an objective approach is required when building the integral index of the quality of life for the subjects of the Russian Federation. The proposed algorithm is resistant to the changes in input data. The method can be used for assessing the quality of life in the territorial units of any level and for calculating integral estimates of the quality change in any poorly formalized system
Keywords
quality of life, development, integral index, measurement error, principal components analysis