Synthesis of a trend’s integral estimate based on a totality of indicators for a time series data

• Main Authors: Alexander Trunov, Volodymyr Beglytsia
• Format: Article
• Language: English
• Published: PC Technology Center 2019-04-01
• Series: Eastern-European Journal of Enterprise Technologies
• Subjects: integral indicator; efficiency of quadratic norm; decomposition norm; constraint on inequality of norm; geometric inequality
• Online Access: http://journals.uran.ua/eejet/article/view/163922

The need for a qualitative estimation of social processes, trends, and activities executed by social governing institutions has been considered. It was determined that the information openness, availability of processing and informational tools create the conditions for efficient administration. An analysis of trends for a time series in the indicators’ dynamics has been performed. The number of priority indicators for analysis and generalization was limited to four, the data were selected from reports by the World Bank, Obozrevatel.ua, and Transparency International. The unified norm has been introduced, which would ensure the comparability of indicators that generate the conclusion on the state, trends, and processes. One of the norms that are applied for Euclidean spaces has been introduced. The system of basic estimation of the integral indicator has been substantiated, based on data about management efficiency, quality of regulation, political stability, absence of violence and CPI index. Expressions for estimation of prognostic values for the integral indicator were derived. It was proposed to use a lower boundary as the estimate to analyze the efficiency of public administration activities. Modeling of the processes of estimation of the integral indicator was performed and a five-year time series for the integral indicator was obtained. The relation between an error of the integral indicator, the size of the sliding window, jumps in first and second derivative from a generalized time series and a permissible error has been represented as a limited inequality. It was proposed to introduce the system of representation and mapping of time series in dimensionless, limited spaces by rotation at an angle around the common axis. The definitions were introduced, the theorem about retaining the local values for relative dimensions and errors were introduced. The influence of the quadratic form on a local relative error of the integral indicator has been shown. The prediction of the development was performed and the assessment of public administration activities was given based on the estimate of the integral indicator