<![CDATA[Negative Binomial Regression is a statistical modeling approach used to analyze count data with overdispersion, where the variance exceeds the mean. This study applies the method to examine the influence of weather factors on the daily number of cyclists crossing the Williamsburg Bridge in New York City. The independent variables used in the analysis include maximum temperature, minimum temperature, and precipitation. The dataset was obtained from the NYC Department of Transportation through the Kaggle platform and covers the period from April 1 to April 30, 2016. The analysis began with a Poisson Regression model; however, the presence of overdispersion was identified, indicated by a high AIC value of 8598.19, suggesting that the model was not suitable. The alternative Negative Binomial Regression model was then employed and produced a significantly lower AIC value of 518.77, demonstrating a superior fit. The findings indicate that maximum temperature has a positive effect on the number of cyclists, while precipitation shows a significant negative effect. Conversely, minimum temperature does not exhibit a meaningful influence. These results highlight the importance of considering weather conditions when planning bicycle-based transportation systems and support the development of sustainable mobility strategies in urban environments.]]>
Copyrights © 2025
