I did this project when I was a grad student in the Master of Urban Spatial Analytics program at University of Pennsylvania. I used R, Python and ArcMap to complete the analysis. Now I put it on a web page. The URL of the report is http://roxyi.github.io/predict_indego/

The purpose of this project is to develop a prediction model that estimates the demand for a new bike-share system in Philadelphia, Indego, by geography. The model helps the Indego bike-share service by predicting the average usage of each bike-share station in a specific time period. In addition, the relationship between bike-share and time is also analyzed in this paper which helps predict when a particular bike-share station will require service. Several factors (e.g., distance to subway stations, percentage of white, labor force) was found to be significantly related to the usage of Indego bike-share in Philadelphia. Peak hour affects bike-share according to the analysis which leads to the finding of daily commuting patterns of Indego bike-share stations. The prediction fit well in both of in-sample prediction and out-of-sample prediction. These findings will help understand the demand for bike-share in Philadelphia and improve the bike-share system including deciding where to expand the bike-share service.