Capital Bike Ubikeshare System In Washington: A Case Study

Econ471 Final Project
Usage Forecast of a city’sbikeshare system in Washington,D.C Courtesy of Capital Bikeshare

Group 4
Xinchen PanDifei Luo Yimeng WuPak Ng
YuxiGuo

University of Illinois at Urbana-Champaign
Dec.02, 2014

Table of Contents

Introduction………………………………………………………. 3
Data analysis……………………………………………………... 4
Making a model/assumption checks……………………………... 8
Conclusion……………………………………………………… 19
Appendix………………………………………………………... 21
References………………………………………………………. 23 Introduction
In this project, we are going to forecast the amount of usage of the bikeshare system in Washington, D.C. Our research question is what are the factors that affect the number of bike rentals. The bikeshare system is a “transportation program, ideal …show more content…
However, this only means that at least one of the variables is significant. Then we need to check which variable is not significant using Anova.

Below is the Anovatable for the regression model.

## Analysis of Variance Table anova(rent) ## Analysis of Variance Table
##
## Response: cnt
## Df Sum Sq Mean Sq F value Pr(>F)
## factor(yr) 1 890908514 890908514 1550.7988 < 2.2e-16 ***
## factor(season) 3 957521736 319173912 555.5840 < 2.2e-16 ***
## factor(mnth) 11 193467220 17587929 30.6152 < 2.2e-16 ***
## factor(holiday) 1 3483977 3483977 6.0645 0.0140313 *
## factor(weekday) 6 14757984 2459664 4.2815 0.0003002 ***
## factor(weathersit) 2 163994925 81997463 142.7325 < 2.2e-16 ***
## I(temp * 41) 1 58065211 58065211 101.0737 < 2.2e-16 ***
## I(100 * hum) 1 6226067 6226067 10.8377 0.0010444 **
## I(windspeed * 67) 1 27703281 27703281 48.2229 8.681e-12 ***
## Residuals 702 403287502 574484
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' '