Comparison of mathematical models of regression analysis on youth crime and delinquency

Abstract

This paper presents a comparative analysis of regression models applied to juvenile crime and minors’ delinquency in the Slovak Republic. The aim of the research is to identify the most appropriate mathematical model for predicting the future development of juvenile crime and minors’ delinquency, based on official statistical data from the Ministry of the Interior of the Slovak Republic for the period 2010–2024. The research analyses linear, exponential, logarithmic, polynomial, and power regression models, and evaluates their reliability using the coefficient of determination (R²). The results show that the logarithmic model is the most suitable for the state and rate of minors’ delinquency (R² = 0.8526; R² = 0.8819), while the polynomial model (R² = 0.981; R² = 0.968) proved to be the most reliable for the state and rate of juvenile crime. An analysis of demographic trends and the crime index reveal a downward trend in both groups, with the prediction models indicating a continuation of this trend. The research provides important information for the development of preventive strategies and policies aimed at reducing youth crime. The research also highlights the fact that, although crime and delinquency show a declining tendency, schools are increasingly facing the emergence of violent socio-pathological phenomena. The findings of this study can serve as a starting dataset for future research on direct criminogenic factors.