This research aims to analyze the perception of logistics operators and retailers regarding freight deliveries in the city of São Paulo. Based on a survey applied to 100 logistics operators and 84 retailers, the data were analyzed by descriptive statistics and multiple correspondence analysis (MCA) was used to investigate the logistics efficiency of off-hours deliveries (OHD) and to indicate issues when carrying out OHD. From that sample, noise appears as the most critical issue of OHD for retailers. From the results, most logistics operators and retailers prefer to deliver cargo at night. The advantages of making OHD are the ability to check/store goods, the accuracy in the delivery schedule due to traffic conditions, and the ease of parking a vehicle to offload goods. Public authorities should improve the infrastructure to receive goods, including public lighting conditions and sidewalks. The correspondence analysis method showed that the level of customer service quality depends on the punctuality of the trucks and the ability to check and store goods. Furthermore, by highlighting the logistics efficiency and issues related to daytime and overnight deliveries by carriers and receivers, it can guide public polices and initiatives of other companies, an aspect that has been lacking in the literature.

Carefully selecting variables in problems with large volumes of data are extremely important, as it reduces the complexity of the model, improves the interpretation of the results, and increases computational efficiency, ensuring more accurate and relevant analyses. This paper presents a comprehensive approach to selecting variables in multiple regression models using the stepwise procedure. As the main contribution of this study, we present the stepwise function implemented in Python to improve the effectiveness of statistical analyses, allowing the intuitive and efficient selection of statistically significant variables. The application of the function is exemplified in a real case study of real estate pricing, validating its effectiveness in improving the fit of regression models. In addition, we presented a methodological framework for treating joint problems in data analysis, such as heteroskedasticity, multicollinearity, and nonadherence of residues to normality. This framework offers a robust computational implementation to mitigate such issues. This study aims to advance the understanding and application of statistical methods in Python, providing valuable tools for researchers, students, and professionals from various areas.