Final Project, Dishaddra and Aishwarya

install.packages(c(tidyverse, sf, tmap, spatstat, maptools, ggplot2, sqldf)) install.packages(“spatstat.geom”, dependencies = TRUE) install.packages(“leaflet.providers”) install.packages(“markdown”, dependencies = TRUE) install.packages(“polyclip”, dependencies = TRUE) install.packages(“ggmap”, dependencies = TRUE)

library(tidyverse) 
library(sf)
library(tmap)
library(spatstat)
library(maptools)
library(sqldf)
library(ggplot2)
library(spatstat.geom)
library(leaflet.providers)
library(markdown)
library(polyclip)
library(ggmap)

# Environment (folder where Chicago Taxi data is stored)
setwd("D:/GaTech/Academic/Fall 2021/Urban Analytics/Final Project")

## We downloaded the data from here: https://data.cityofchicago.org/Transportation/Taxi-Trips/wrvz-psew
## Some code taken from: http://lab.rady.ucsd.edu/sawtooth/business_analytics_in_r/maps.html
## Thanks to Prof. Clio Andris for her teachings and help.

# Read data

TT2019 <- read.csv("Taxi_Trips_-_2019.csv")


TT2020 <- read.csv("Taxi_Trips_-_2020.csv")

# Extract month from Timestamp 

TT2019month <-TT2019%>%
 mutate(month = substring(TT2019$Trip.Start.Timestamp, 1,2))



TT2020month <-TT2020%>%
 mutate(month = substring(TT2020$Trip.Start.Timestamp, 1,2))

# Aggregate months

tripsBymonth2019 <- TT2019month  %>%
  group_by(Pickup.Community.Area, month) %>%
  summarise(trips=n())
 
tripsBymonth2020 <- TT2020month  %>%
  group_by(Pickup.Community.Area, month) %>%
  summarise(trips=n())

library(reshape2)
TripByCommunityAreaAndMonth2019 <- dcast(unique(tripsBymonth2019), Pickup.Community.Area ~ month, value.var="trips")
TripByCommunityAreaAndMonth2020 <- dcast(unique(tripsBymonth2020), Pickup.Community.Area ~ month, value.var="trips")

# Merge two years to find difference


fulldata <- merge(TripByCommunityAreaAndMonth2019, TripByCommunityAreaAndMonth2020, by.x = "Pickup.Community.Area",
                   by.y = "Pickup.Community.Area",
                   suffixes = c("y2019","y2020"))

# Format Table

res <- fulldata[, grepl("y2019", colnames(fulldata))] - fulldata[, grepl("y2020", colnames(fulldata))]

colnames(res) <- paste(colnames(fulldata[, grepl("y2019", colnames(fulldata))]),
                       colnames(fulldata[, grepl("y2020", colnames(fulldata))]), sep = "_")

Final <- cbind(fulldata,res)

write.csv(Final,"D:/GaTech/Academic/Fall 2021/Urban Analytics/Final Project/Final.csv")

#We performed spatial joins on GIS to find interrelationships with CTA railway stations and Community area wise Hardship indices.
#We converted presence or absence of CTA railway station in a community area into a binary field.
#Similarly we bifurcated Hardship indices that range from 0 to 100 into a binary assigning 0 for under 50 and 1 for over and equal to 50.

Final_Table <- read.csv("Hardship_Index.csv")
Final_Table$Nomalisedapril <- Final_Table[,c(30)]/Final_Table[,c(6)]

# Anoval Test for Presence of Public transport (for April)

one.way <- aov(Final_Table[,c(60)] ~ Final_Table[,c(42)], data = Final_Table)

summary(one.way)

# Anoval Test for Hardship Index (for April)



one.way <- aov(Final_Table[,c(60)] ~ Final_Table[,c(43)], data = Final_Table)

summary(one.way)