##### Lesson 5: Let us 'R'eview #####

# Setting the working directory 
setwd(“provide the path to your folder“)
list.files()

# Read the datafile #
naturalgas_data = read.csv("Natural_Gas_Consumption_by_ZIP_Code_-_2010.csv",header=T)

# extract subset of the data #
index1 = which(naturalgas_data$Building.type..service.class == "Large residential")
index2 = which(naturalgas_data$Building.type..service.class == "Commercial")
index3 = which(naturalgas_data$Building.type..service.class == "Institutional")

Residential = naturalgas_data$Consumption..GJ.[index1]
Commercial = naturalgas_data$Consumption..GJ.[index2]
Institutional = naturalgas_data$Consumption..GJ.[index3]

# Order statistics and plots #
stripchart(Residential,font=2,pch=21,cex=1,xlab="Consumption in GJ",font.lab=2)
boxplot(Residential,horizontal=T,col="grey",add=T)

# histogram #
hist(Residential,font=2,font.lab=2,xlab="Consumption in GJ",col="grey")

# Summary statistics #

# average #
mean(Residential,na.rm=T)

# median #
median(Residential,na.rm=T)
quantile(Residential,na.rm=T,0.5)

# variance and standard deviation #
var(Residential,na.rm=T)
sd(Residential,na.rm=T)

# skewness #
library(moments)
skewness(Residential,na.rm=T)

# quantiles and summary #
quantile(Residential,na.rm=T)

summary(Residential)

# comparing the data # 
all_data = cbind(Residential,Commercial, Institutional)
boxplot(all_data,horizontal=T)

cv_residential = (sd(Residential,na.rm=T)/mean(Residential,na.rm=T))*100
cv_commercial = (sd(Commercial,na.rm=T)/mean(Commercial,na.rm=T))*100
cv_institutional = (sd(Institutional,na.rm=T)/mean(Institutional,na.rm=T))*100