-04-visualization

CANVAS:

#####################
#PACKAGE PREREQUISITE
#####################
install.packages("tidyverse")
library(tidyverse) 

mpg
View(mpg)

#Type View(mpg) in the console for a spreadsheet view of the data
View(mpg)

####
#YOUR TURN! & SOLUTION
####
# additional documentation for the mpg data set 
?mpg

####
#CANVAS
####

#LET’S CREATE OUR “CANVAS”

# left 
ggplot (data = mpg)
# right 
ggplot(data = mpg, aes(x = displ, y = hwy))

GEOMS:

#######
#GEOMS
#######
#• We display data with geometric shapes 
#• ~ 30 built-in geoms (with many more offered by other pkgs)
#https://ggplot2.tidyverse.org/reference/
#Type geom_ + tab in the console

####
#UNIVARIATE GEOMS
####
ggplot(data = mpg, aes(x = hwy)) +  
  geom_histogram() 

ggplot(data = mpg, aes(x = hwy)) +  
  geom_freqpoly() 

ggplot(data = mpg, aes(x = hwy)) +  
  geom_density()

ggplot(data = mpg, aes(x = class)) +  
  geom_bar()

#ggplot(data = mpg, aes(x = class)) +  geom_bar()
# • This is called an aesthetic mapping argument 
# • Every geom requires a mapping argument 
# • Some geoms require just one (x variable) 
# • While other geoms require two (x & y variable)

####
#BIVARIATE GEOMS
####
ggplot(data = mpg, aes(x = displ, y = hwy)) +  
  geom_point() 

ggplot(data = mpg, aes(x = class, y = hwy)) +  
  geom_boxplot() 

ggplot(data = mpg, aes(x = class, y = hwy)) +  
  geom_violin()
  
####
#YOUR TURN! & SOLUTION
####
#1: import data 
customer<-read_csv("data/CustomerData.csv") 
#2: distribution of DebtToIncomeRatio variable 
ggplot(data = customer, aes(x = DebtToIncomeRatio)) + geom_histogram() 
#3: distribution of JobCategory variable 
ggplot(data = customer, aes(x = JobCategory)) + geom_bar() 
#4: scatter plot for HHIncome vs CardSpendMonth 
ggplot(data = customer, aes(x = HHIncome, y = CardSpendMonth)) + geom_point()

####
#NON-MAPPING AESTHETICS 
####
# We can also change other visual aesthetics in our graphics 
# • color 
# •size 
# • sh△pe (0-25 ?pch) 

ggplot(data = mpg, aes(x = displ, y = hwy)) +  
  geom_point(color = "blue", size = 2, shape = 17, alpha = .5)

# Why are some points darker than others? 
# • Try geom_jitter in place of geom_point 
# • What do you think geom_jitter does?

####
#ADDING A 3RD DIMENSION
####

ggplot(data = mpg, aes(x = displ, y = hwy)) +  
  geom_point(color = "blue")

#By moving the color argument to within aes(), 
#we can map a 3rd variable to our plot 
ggplot(data = mpg, aes(x = displ, y = hwy, color = class)) +  
  geom_point()

#A common error…what happened??? 
ggplot(data = mpg, aes(x = displ, y = hwy, color = "red")) +  
  geom_point()
  
####
#YOUR TURN! & SOLUTION
####
#1. Create a scatter plot of HHIncome	vs	CardSpendMonth 
#and color all points blue. 
ggplot(customer, aes(x = HHIncome, y = CardSpendMonth)) +   
  geom_point(color = "blue")

#2. Create a scatter plot of HHIncome	vs	CardSpendMonth 
#and color all points based on #   whether or not the customer is retired.. 
ggplot(customer, aes(x = HHIncome, y = CardSpendMonth, color = Retired)) +   
   geom_point()

FACET:

########
#FACETS
########
#FACETS = SMALL MULTIPLES

#• The facet functions provide a simple way to create small multiples 
#• facet_wrap:  primarily used to create small multiples based on a single variable 

ggplot(data = mpg, aes(x = displ, y = hwy)) +  
  geom_point() +   
  facet_wrap(~class, nrow = 2)

#• use nrow or ncol to specify dimensions 
#• ?facet_wrap to see other arguments to control the output

#• facet_grid:  primarily used to create a small multiples grid based on two variables

ggplot(data = mpg, aes(x = displ, y = hwy)) +   
  geom_point() +   facet_grid(drv ~ cyl)

####
#YOUR TURN! & SoLution
####

# 1. Create a scatter plot of HHIncome vs CardSpendMonth 
#facetted by JobCategory & Gender. 
ggplot(customer, aes(x = HHIncome, y = CardSpendMonth)) +   
  geom_point() +   
  facet_wrap(~ JobCategory)

# 2. Create a scatter plot of HHIncome vs CardSpendMonth 
#facetted by JobCategory & Gender. 
ggplot(customer, aes(x = HHIncome, y = CardSpendMonth)) +   
  geom_point() +   
  facet_grid(Gender ~ JobCategory)

# 3. Assess UnionMember across each JobCategory 
ggplot(customer, aes(x = UnionMember)) +   
  geom_bar() +   
  facet_wrap(~ JobCategory)

TITLES & AXES:

################
#TITLES & AXES:
################
####
#ADDING TITLES
####

# top 
ggplot(data = mpg, aes(x = class, y = hwy)) +   
geom_point() +  
ggtitle("Displacement vs Highway MPG") 

# bottom 
ggplot(data = mpg, aes(x = class, y = hwy)) +   
  geom_boxplot() +  
  ggtitle("Displacement vs Highway MPG”,subtitle = “Data from 1999 & 2008")

####
#ADJUSTING AXIS SCALES
####

ggplot(data = txhousing, aes(x = volume, y = median)) +   
  geom_point(alpha = .25) +  
  scale_x_log10()

####
#ADJUSTING AXIS TITLES & LABELS
####
ggplot(data = txhousing, aes(x = volume, y = median)) +   
  geom_point(alpha = .25)  +  
  scale_y_continuous(name = "Median Sales Price", labels = scales::dollar) +  
  scale_x_log10(name = "Total Sales Volume", labels = scales::comma)

####
#PUT IT ALL TOGETHER
####

ggplot(data = txhousing, aes(x = volume, y = median)) +  
  geom_point(alpha = .25) +  
  scale_y_continuous(name = "Median Sales Price", labels = scales::dollar) +  
  scale_x_log10(name = "Total Sales Volume", labels = scales::comma) +  
  ggtitle("Texas Housing Sales",subtitle = "Sales data from 2000-2010 provided by the TAMU real estate center")

####
#YOUR TURN! & SOLUTION
####
#1. Remove all missing values from the customer data and then… 
#2. Create a scatter plot of HHIncome vs CardSpendMonth facetted by JobCategory and… 
#3. add a title, subtitle, and nicely format the axes.

customer %>%  
  na.omit() %>%  
  ggplot(aes(x = HHIncome, y = CardSpendMonth)) +   
  geom_point(alpha = .3) +   
  facet_wrap(~ JobCategory) +  
  scale_x_log10("Household Income", labels = scales::dollar) +  
  scale_y_log10("Monthly Card Expenditures", labels = scales::dollar) +  
  ggtitle("Relationship between income and credit card expenditures",subtitle = "All customers (4th Qtr 2015)")

OVERPLOTTING:

#############
#OVERPLOTTING
#############

####
#LAYERING HELPS DISPLAY PATTERNS 
####

ggplot(data = txhousing, aes(x = volume, y = median)) +   
  geom_point(alpha = .25)  +  
  scale_x_log10() +  
  geom_smooth()

ggplot(data = txhousing, aes(x = volume, y = median)) +   
  geom_point(alpha = .25)  +  
  scale_x_log10() +  
  geom_smooth(method = "lm")

ggplot(data = txhousing, aes(x = volume, y = median)) +   
  geom_point(alpha = .25)  +  
  scale_x_log10() +  
  geom_smooth(method = "lm") +  
  facet_wrap(~ month)

#######################
#YOUR TURN! & SOLUTION
######################
# 1. Remove all missing values from the customer data and then… 
# 2. Create a scatter plot of HHIncome vs CardSpendMonth facetted by JobCategory and… 
# 3. add a title, subtitle, and nicely format the axes and… 
# 4. add a linear line to assess if the slope changes across JobCategory

customer %>%  
  na.omit() %>%  
  ggplot(aes(x = HHIncome, y = CardSpendMonth)) +   
  geom_point(alpha = .3) +   
  geom_smooth(method = "lm") +  
  facet_wrap(~ JobCategory) +  
  scale_x_log10("Household Income", labels = scales::dollar) +  
  scale_y_log10("Monthly Card Expenditures", labels = scales::dollar) +  
  ggtitle("Relationship between income and credit card expenditures", subtitle = "All customers (4th Qtr 2015)")

FUNCTIONS TO REMEMBER:

ggplot() Initializes a ggplot object (creates the blank canvas) 
aes() Creates aesthetic mappings 
geom_xx Geometric shapes to plot the data 
color, shape, size, alpha, etc Aesthetic parameters 
facet_wrap, facet_grid Create small multiples position 
Position argument (primarily used with bar charts) 
coord_xx Functions to adjust the coordinate system 
scale_xx Functions to adjust x and y axis