Thetazero Pubs
I1400094

How does Bike Sharing Demand Change?

Keiya Uchida
December 16, 2015DATA_ADSENSE

Introduction

Method

  • Summary of the dataset (rattle)
  • Correlation plot (rattle)
  • Linear regression analysis - temp v.s. atemp
  • Decision tree (rattle)
  • Plot- three important variables
  • Plot- working day v.s. non- working day
  • Multiple linear regression

Summary of the dataset (rattle)

Data Fields

  • 12 VARIABLES (datetime, season, holiday, workingday, weather, temp, atemp, humidity, windspeed, casual, registered, count)

  • holiday - whether the day is considered a holiday

  • workingday - whether the day is neither a weekend nor holiday

  • temp - temperature in Celsius DATA_ADSENSE

  • atemp - ?feels like? temperature in Celsius

Correlation plot (rattle)

alt text

Linear regression analysis - temp v.s. atemp

train <- read.csv("~/Desktop/Kaggle/train (4).csv")
reg_temp <- lm(formula = count ~ temp, data = train)

Multiple R-squared: 0.1556, Adjusted R-squared: 0.1555

reg_atemp <- lm(formula = count ~ atemp, data = train)

Multiple R-squared: 0.1519, Adjusted R-squared: 0.1519

Decision tree (rattle)

alt text

Plot- three important variables

# tempture
plot(train$temp, train$count)
abline(reg_temp)

plot of chunk unnamed-chunk-3

Plot- three important variables

plot of chunk unnamed-chunk-4

Plot- three important variables

plot of chunk unnamed-chunk-5

Plot- working day v.s. non- working day

alt text

Plot- working day v.s. non- working day

alt text

Multiple linear regression

Copyright © 2016 thetazero.com All Rights Reserved. Privacy Policy