Bike Renting Data Analysis
A team project that analyzes a dataset about bike renting service in Seoul.
Description
The purpose of this project is to research on how the weather condition such as humidity, windspeed, visibility, temperature, dewpoint temperature influence the bike rental amount in different hours through out the day. The main research question is whether the weather condition through out the year will influence the bike rental amount. Therefore, the main effect model and main effect interaction are compared to see whether the weather condition can significantly influence the rental bike amount.
Features
Analyze explanatory variables and response variable by different methods to find correlations between them.
Build regression models for dataset and use stepAIC and other methods to choose the best model.
Box-Cox transformation on response variable to improve the model in terms of R-squared.
Applied model diagnostics to deal with outliers and leverage.
Model building process is saved in R Markdown file.
A detailed report is written to show final results:
(A part of the report. The complete report could be found on Github)