Статистичний аналіз динаміки кількості вчинених злочинів на території України у 1990-2020 роках

Abstract

The paper reveals an information-analytical study of the dynamic range of the total number of crimes committed on the territory of Ukraine in the period from 1990 to 2020, and the corresponding indicator of the number of convicted people. Calculations based on the Irwin criterion indicate an abnormal increase in crime in 2003 and 2013 and an abnormal decrease in 2020. At the same time, there are no corresponding atypical values in the number of convicted persons in Ukraine. Since the detected uncharacteristic increase in the number of crimes is observed precisely in the years preceding the Orange Revolution (2004) and the Revolution of Dignity (2014), their appearance may indicate the growth of protest sentiments in society and bear a certain indicator character of popular opposition to the actions of the current government. Instead, the detected anomalous decrease in the number of crimes in 2020 may be a consequence of the general trend of decreasing crime due to the introduction of restrictive quarantine measures caused by the COVID-19 pandemic. In order to test the hypothesis about the presence of a trend in the studied crime indicator, the criterion of “increasing” and “declining” series and the criterion of series based on the use of the median are used in the work. However, the calculations showed that, despite the presence of a weak trend, after eliminating abnormal levels and smoothing a number of dynamics, it is not possible to obtain an adequate model of the growth curve of the number of crimes and conduct extrapolation forecasting on its basis. The empirical data of the studied time series can best be approximated only with the help of polynomials higher than the third degree. However, in this case, the obtained approximating functions will reflect random deviations and will be unacceptable for forecasting the future level of the number of crimes in Ukraine. Therefore, the construction of prognostic functions of crime involves the further development of an autoregressive integrated moving average model.