In this notebook, we analyze data on taxi trips in New York City to gain insights into how different factors affect trip fares. We start by visualizing the spatial distribution of trip origins and the relationship between fare and distance. Then, we fit a regression tree and a random forest to predict trip fares based on variables such as pickup location, time of day, day of the week, and month. We compare the performance of the two methods and highlight the most important predictors. Finally, we visualize the predicted fares and explore how they vary across the city.