Kyle_Walter_HW7.Rmd

---
title: "Week 7 Homework"
author: "Kyle Walter"
date: "11/15/2020"
output:
  word_document: default
  html_document: default
---

IST687	– Viz	Map HW:	Median	Income
Download	the	dataset	from	the	LMS that	has	median	income	by	zip	code	(an	excel	file).
Step	1: Load	the	Data
1) Read	the	data	– using	the	gdata	package	we	have	previously	used.			

```{r}
require(gdata)
require(readxl)
zipData <- read_xlsx("J:/IST687 - Into to Data Science/Week 7/MedianZIP_2_2.xlsx")
colnames(zipData) <- c("zip","median","mean","population")
```

2) Clean	up	the	dataframe
a. Remove	any	info	at	the	front	of	the	file	that’s	not	needed
b. Update	the	column	names	(zip,	median,	mean,	population)
```{r}
zipData <- zipData[-1,]
tail(zipData) #looks like this good data
zipData$median <- gsub(",","",zipData$median)
zipData$mean <- gsub(",","", zipData$mean)
zipData$population <- gsub(",","",zipData$population)
zipData$mean <- ifelse(zipData$mean==".",0,zipData$mean)
zipData$median <- as.numeric(zipData$median)
zipData$mean <- as.numeric(zipData$mean)
zipData$population <- as.numeric(zipData$population)
zipData$zip <- as.character(sprintf("%05d",as.numeric(zipData$zip)))
```


3) Load	the	‘zipcode’	package

```{r}
ZipPackage <- "http://cran.r-project.org/src/contrib/Archive/zipcode/zipcode_1.0.tar.gz"
install.packages(ZipPackage, repos = NULL, type = "source")
require(zipcode)
data(zipcode)
```


4) Merge	the	zip	code	information	from	the	two	data	frames	(merge	into	one	
dataframe)

```{r}
require(dplyr)
Combine.df <- inner_join(zipData, zipcode, by = c("zip"="zip"))
```


5) Remove	Hawaii	and	Alaska	(just	focus	on	the	‘lower 48’	states)

```{r}
Combine.df <- Combine.df[which(Combine.df$state!="HI"&Combine.df$state!="AK"),]
```


Step	2: Show	the	income	&	population	per	state
1) Create	a	simpler	dataframe,	with	just	the	average	median	income	and	the	the	
population	for	each	state.

```{r}
Income <- tapply(Combine.df$median, Combine.df$state, mean)
states <- rownames(Income)
StateIncome <- data.frame(states, Income)
population <- tapply(Combine.df$population, Combine.df$state, sum)
state <- rownames(population)
StatePop <- data.frame(state, population)
Simple.DF <- inner_join(StateIncome, StatePop, by = c("states"="state"))
```


2) Add	the	state	abbreviations	and	the	state	names	as	new	columns	(make	sure	the	
state	names	are	all	lower	case)
```{r}
states <- data.frame(state.abb, state.name)
Simple.DF <- inner_join(Simple.DF, states, by = c("states"="state.abb"))
Simple.DF$stateName <- tolower(Simple.DF$state.name)
```

3) Show	the	U.S.	map,	representing	the	color	with	the	average	median	income	of	that	
state

```{r}
require(ggplot2)
require(mapproj)
Simple.DF <- Simple.DF[order(Simple.DF$Income),]
us <- map_data("state")
income.map <- ggplot(data = Simple.DF, aes(map_id = stateName))
income.map <- income.map+geom_map(map = us, aes(fill = Income), color="black")
income.map  <- income.map+expand_limits(x = us$long, y = us$lat)+coord_map()+ggtitle("Average income by State")
income.map
```


4) Create	a	second	map	with	color	representing	the	population	of	the	state
```{r}
pop.map <- ggplot(Simple.DF, aes(map_id = stateName))
pop.map <- pop.map + geom_map(map=us, aes(fill = population), color="black")
pop.map <- pop.map + expand_limits(x=us$long, y=us$lat)+coord_map()+ggtitle("Population by State")
pop.map
```

Step	3:	Show	the	income	per	zip	code
1) Have	draw	each	zip	code	on	the	map,	 where	the	color	of	the	‘dot’	is	based	on	the	
median	income.	To	make	the	map	look	appealing,	have	the	background	of	the	map	
be	black.

```{r}
Combine.df <- left_join(Combine.df, states, by = c("state"="state.abb"))
Combine.df$state.name <- tolower(Combine.df$state.name)
zip.income <- ggplot(Combine.df, aes(map_id = state.name))
zip.income <- zip.income+geom_map(map = us, color="black")+geom_point(Combine.df, mapping = aes(x=longitude, y=latitude, color = median))
zip.income <- zip.income + expand_limits(x=us$long, y = us$lat)
zip.income <- zip.income+coord_map()+ggtitle("Income by Zip")
zip.income
```


Step	4:	Show	Zip	Code	Density
1) Now	generate	a	different	map,	one	where	we	can	easily	see	where	there	are	lots	of	
zip	codes,	and	where	there	are	few	(using	the	‘stat_density2d’	function).

```{r}
zip.density <- ggplot(Combine.df, aes(x=longitude, y=latitude))
zip.density <- zip.density+geom_point(color="red", alpha = .6)+stat_density2d()
zip.density <- zip.density+expand_limits(x=us$long, y=us$lat)+coord_map()
zip.density
```


Step	5:	Zoom	in	to	the	region	around	NYC
1) Repeat	steps	3	&	4,	but	have	the	image	/	map	be	of	the	northeast	U.S.	(centered	
around	New	York).


```{r}
library(tmaptools)
Home <- geocode_OSM("New York, NY")
Home <- data.frame(Home[[2]])
zoomAmount <- 4
ylim <- c(Home[2,]-zoomAmount, Home[2,]+zoomAmount)
xlim <- c(Home[1,]-zoomAmount-1, Home[1,]+zoomAmount+1)
NEZips <- Combine.df
NEZips <- NEZips[NEZips$longitude>xlim[1],]
NEZips <- NEZips[NEZips$longitude<xlim[2],]
NEZips <- NEZips[NEZips$latitude>ylim[1],]
NEZips <- NEZips[NEZips$latitude<ylim[2],]
CapofWorld <- ggplot(NEZips, aes(map_id = state.name))
CapofWorld <- CapofWorld+geom_map(map=us, color = "white")+geom_point(NEZips, mapping = aes(x=longitude, y=latitude, fill = median, color = median))
CapofWorld <- CapofWorld+expand_limits(x=xlim, y=ylim)+coord_map()+ggtitle("Median North Easter Income by Zip")
DensofWorld <- ggplot(NEZips, aes(x=longitude, y=latitude))+geom_point(color="red", alpha = .6)+stat_density2d()
DensofWorld <- DensofWorld+expand_limits(x=xlim, y=ylim)+coord_map()+ggtitle("North East US Zip Code Density")
CapofWorld
DensofWorld
```