Lab3.Rmd

---
title: 'L3: ESDA in R'
author: "Chad Fisher"
date: "`r Sys.Date()`"
output: html_document
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
```

# Read in Data

```{r Load_data_and_rename_variables}
setwd("C:/Users/Chad/Box/UEP236_Labs")

library(sf)

towns_Mass <- read_sf('./Lab3/Towns_MA_spjoin.shp')

#Change Names
names(towns_Mass)[13]<- paste("Population")
names(towns_Mass)[14]<- paste("Population Density")
names(towns_Mass)[17]<- paste("Black Population")
names(towns_Mass)[32]<- paste("Hispanic Population")
names(towns_Mass)[36]<- paste("Income")
names(towns_Mass)[37]<- paste("Rent")
names(towns_Mass)[33]<- paste("Workers")
names(towns_Mass)[35]<- paste("Unemployed Workers")
names(towns_Mass)[46]<- paste("Persons Poverty Level")
names(towns_Mass)[45]<- paste("Total Persons Poverty Determined")
names(towns_Mass)[52]<- paste("Housing Value")
names(towns_Mass)[66]<- paste("Total Releases")
names(towns_Mass)[64]<- paste("Onsite Releases")
names(towns_Mass)[65]<- paste("Offsite Releases")
```

```{r Calculate_Variables}
#calculate poverty proportion
towns_Mass$PovProp <- towns_Mass['Persons Poverty Level']/towns_Mass['Total Persons Poverty Determined']
#calculate percent Black and percent Hispanic
towns_Mass$HispPerc <- (towns_Mass['Hispanic Population']/towns_Mass['Population'])*100
towns_Mass$BlackPerc <- (towns_Mass['Black Population']/towns_Mass['Population'])*100
#calculate unemployment percent
towns_Mass$UnempPerc <- (towns_Mass['Unemployed Workers']/towns_Mass['Workers'])*100

library(Hmisc)
towns_Mass$IncCat<- cut2(towns_Mass$Income, g=4, levels.mean = TRUE)
summary.factor(towns_Mass$IncCat)
```

```{r data_subset_and_summary}
towns_Mass_subset <- towns_Mass[, c("Population", "Population Density", "PovProp","HispPerc", "BlackPerc", "UnempPerc", "Housing Value", "Rent", "Income", "Total Releases","Onsite Releases", "Offsite Releases")]

towns_Mass_df <- data.frame(towns_Mass_subset)

library(stargazer)

#text file saved to table1.txt
stargazer(towns_Mass_df, median=TRUE, type = "text", title="Descriptive statistics", digits=1, out="tablefinal.txt")

#html file which looks nicer
stargazer(towns_Mass_df, median=TRUE, type = "html", title="Descriptive statistics", digits=1, out="table1.html")

my_subset <- towns_Mass_df[,c('Rent','Population.Density','Income','Offsite.Releases')]

stargazer(my_subset, median=TRUE, type = "html", title="Descriptive statistics", digits=1, out="my_table.html")

```

# Making Plots with ggplot2

```{r ggplot}
library(ggplot2)

ggplot(my_subset,aes(x=Rent)) +
  geom_histogram(bins=50) +
  ggtitle("Histogram of Rent from TRI sites by town") +
  xlab("Rent") +
  ylab("Count of Towns")
```

```{r boxplot}
p <- ggplot(towns_Mass, aes(x = IncCat, y = Rent)) +
  stat_boxplot(geom = "errorbar", # Error bars
               width = 0.25) + # Bars width
  ggtitle("Boxplots of Rent from TRI sites by
Income Category") +
  xlab("Income Category") +
  ylab("Rent")+
  geom_boxplot()

p
```

```{r Parallel_coordinate_plots}
library(GGally)
ggparcoord(data= towns_Mass_subset,
           columns = c(10:12),
           groupColumn = 9,
           scale = "std",
           #boxplot=TRUE,
           showPoints = TRUE,
           title="Parallel coordinate plot of Releases grouped by standardized income",
           mapping = ggplot2::aes(size = 1)) +
  ggplot2::scale_size_identity() + xlab("") +
  ylab("Income") +
  scale_color_gradient(low = "yellow",
                       high = "brown",
                       aesthetics = "colour") +
  theme_dark()

library(RColorBrewer)
display.brewer.all()
display.brewer.pal(7, "Spectral")

ggparcoord(data= towns_Mass_subset,
           columns = c(10:12),
           groupColumn = 9,
           scale = "std",
           #boxplot=TRUE,
           showPoints = TRUE,
           title="Parallel coordinate plot of Releases grouped by standardized income",
           mapping = ggplot2::aes(size = 1)) +
  ggplot2::scale_size_identity() + xlab("") +
  ylab("Income") +
  scale_color_gradientn(colors = brewer.pal(7,"Spectral")) +
  theme_dark()

```

```{r scatterplot_matrix}
ggpairs(towns_Mass_subset, columns = c(8, 2, 9, 10, 12),
        lower = list(continuous = "smooth"),
        title = "Scatter Plot Matrix", axisLabels = "show")
```