UDC: 331.14-021.111:330.4:004
https://doi.org/10.25198/2077-7175-2019-5-31
RELIABILITY OF STATISTICAL INDICATORS IN THE METHODOLOGY OF MEASURING «DIGITAL ECONOMY»
V. N. Afanasiev
Orenburg state University, Orenburg, Russia
e-mail: kstat@mail.osu.ru
Abstract. The purpose of the article is the possible reasons for the violation of the reliability of statistical indicators in the “digital economy”, ultimately violating the effectiveness of management decisions, leading to their inefficiency in conditions of intensive information flows. So far, the errors associated with the construction of indicators in the economy have not been sufficiently investigated.
The main results of the study are theoretical descriptions of the methodological foundations of errors that violate the reliability of statistical indicators in the “digital economy”.
By the nature of the action, errors are divided into systematic, random, and, if possible, predictions - predictable in magnitude and probability of occurrence and unpredictable (uncertain) errors.
Systematic errors are the most dangerous. Their source is incorrect principles and methods of calculation. Therefore, such errors are difficult to eliminate.
Random errors are in the nature of random variables or random functions, i.e., the limits within which their size can be found, can be argued only with some probability.
Often the method of calculation does not take into account the inevitability of variation of indicators, limited to a deterministic description of economic phenomena. If the objective variation is studied, in some cases its estimation is made approximately. This is due to the lack of mass implementation of processes, schematization of random processes on the basis of hypotheses about their stationarity, linearity; conditional choice of the type of distribution of random variables, etc.
Any calculation takes into account only the essential factors of the phenomena, keeping the possibility of residual variation due to unaccounted factors. The actual data are formed as a result of the influence of all the factors that determined this or that phenomenon, so the probabilistic nature of the implementation and regulatory calculations should be taken into account.
Random error occurs when aggregating and aggregating indicators due to the random nature of the shifts in the population. Probabilistic characteristics determined on average for the whole unit by mass of cases and over a long period are unreliable indicators for the calculation of individual parts of the unit (for example, for individual enterprises, activities).
The study made it possible to draw the following conclusions. Any deviation from the required reliability of indicators associated with economic damage. Insufficient accuracy of calculations can lead to imbalances in the economy of organizations, regions, States, economic communities (with the globalization of the economy). In these circumstances, the creation of reliable statistical indicators is necessary to improve the practice of calculations at various levels of the “digital economy”, as well as for the further development of statistical science.
Keywords: statistical indicator, systematic and random errors, possible causes of errors in statistical indicators in the «digital economy».
