This paper analyzes traffic accidents in Honduras, examining trends, underlying causes, and recommendations to improve traffic safety in the country. A literature review is carried out on traffic accidents and country’s statistical data is presented, as well as the design and implementation of a LSTM neural network model to generate predictions of traffic accidents, on a monthly and weekly time series. The model results are different; weekly time series offers better results due a larger set of data to learn from, 212 data points versus 48. Presented in this paper is the prediction of traffic accidents in the department of Cortés (selected due to its higher accident rate), however, access is given to the model through a Google Colab file, available to any user and/or interested authorities. This paper serves as a starting point to improve statistical analysis and generate data-driven solutions to the current state of traffic accidents in Honduras.