Статистичний аналіз і прогнозування динаміки ДТП з постраждалими в Дніпропетровській області

Purpose. The study aims to identify seasonal and weather-related factors influencing the number of road traffic accidents (RTAs) with casualties in Dnipropetrovsk Oblast, in order to provide a basis for short-term accident forecasting and improve the effectiveness of traffic safety prevention measures. Design / Method / Approach. The study is based on monthly statistical data from 2020 to 2025, sourced from the Patrol Police of Ukraine and the Main Department of Statistics in Dnipropetrovsk Oblast. The analysis integrates boxplot visualization, one- and two-way ANOVA, correlation analysis, and multiple regression analysis, accounting for temperature, precipitation, humidity, wind speed, and calendar month. Findings. A statistically significant impact of the calendar month was confirmed as the main seasonal factor. The regression model showed a clear link between RTAs and combined weather and calendar variables. Accident rates peaked in summer–autumn and dropped in winter–spring, with February showing the least variation. Monthly factors proved useful for short-term forecasting. Time series analysis enabled tracking trends (2020–2024) and projecting them through 2026. Theoretical Implications. The results enhance the understanding of the interplay between seasonal, calendar, and weather-related factors and accident rates within a regional context. Practical Implications. The findings can be used by local authorities, police, and road safety services to plan preventive measures during periods of increased accident risk. Originality / Value. A comprehensive approach is proposed for analyzing RTAs at the regional level, involving various types of statistical analysis and seasonal forecasting. The methodology can be adapted for other regions of Ukraine. Research Limitations / Future Research. The study is limited to the 2020–2025 period and does not account for social, behavioral, or infrastructure-related factors. Future research should incorporate additional explanatory variables and apply multifactor forecasting methods. Article Type. Applied Research. PURL: https://purl.org/cims/4.311