Bike-Sharing-Kaggle-

The purpose of this study is to understand what influences bike rental usage and also predict it in order to satisfy demand. This study is based on regression analysis methods with aim to create a prediction model for bike usage. We will use a dataset from Capital Bikeshare system, Washington D.C., USA which is publicly available in http://capitalbikeshare.com/system-data. This database was aggregated with another database with weather and seasonal information from http://www.freemeteo.com. For our analysis purpose we will use a data sample with 1500 observations for model training and another smaller sample with 500 observations for model evaluation.