BINARY LOGISTIC REGRESSION ANALYSIS OF TRAFFIC ACCIDENT RISK FACTORS IN INDONESIA

Abstract

This study analyzes the factors influencing the severity of traffic accidents in Indonesia using binary logistic regression. Data from Kaggle includes variables such as age, gender, driving experience, lighting conditions, and weather conditions. The results indicate that poor lighting and adverse weather significantly increase the likelihood of fatal accidents by 88.8% and 96.5%, respectively. The logistic regression model achieves 76% accuracy with a good data fit (Hosmer-Lemeshow p-value = 0.144). These findings provide valuable insights for traffic safety policies and infrastructure development.