Spatial analysis of crime at the level of detailed data of the Police of the Czech Republic

This bachelor's thesis focuses on the analysis of property crime across districts of the Czech Republic during the period 2016–2024. The main objective is to assess the impact of the COVID-19 pandemic on the development of property crime and to identify which socio-economic factors were statistically significantly associated with it. The analysis utilizes panel data that combine spatial (district-level) and temporal (yearly) dimensions, and applies three econometric models: Pooled OLS, Fixed Effects, and Random Effects. Based on the results of various statistical tests (F-test, Hausman test, Breusch-Pagan test, and Wooldridge test), the Fixed Effects model was identified as the most suitable. This model allowed for the control of unobserved heterogeneity across districts and provided consistent estimates. Statistically significant factors influencing the level of property crime included, for instance, the average wages and COVID period. Conversely, the pandemic period was associated with a decrease in crime in some models, likely reflecting changes in mobility and social behavior. The analysis also included spatial diagnostics using the Local Moran's I, which revealed clusters of higher crime, particularly in urban regions. The findings contribute to a deeper understanding of the regional and temporal patterns of crime and can serve as a basis for more effective security and social policy planning.