intro to GG plot with assignment

intro_ggplot.R

#install.packages(c('datasauRus',"ggplot2","dplyr"))
library(ggplot2)
library(datasauRus)
library(dplyr)

#Showing the importance of data visualization
# The anscombe dataset has four x and four y variables 
data("anscombe")
anscombe

# We can create summary statistics about each pair

apply(anscombe,2,mean)
apply(anscombe,2,sd)
anscombe.cor = c(0,0,0,0)
for(i in 1:4){
  anscombe.cor[i] = cor(anscombe[,i],anscombe[,i+4])
}
anscombe.cor

# Each of the pairs of variables has nearly identical means, standard deviations, and correlations
# But there is something different about these four pairs of data

par(mfrow=c(2,2))
ggplot(anscombe,aes(x=x1,y=y1)) + geom_point() + expand_limits(x = 0, y = 0) + geom_smooth(method="lm")

# It was not until we plotted the data that we see any problem with it. 
# This illustrates the importance of "exploratory visualization" to data science
# You can't always rely solely on summary statistics and fitting models!

# The code above uses the package ggplot2 to visualize the data
# ggplot2 uses the "grammar of graphics" to build plots for both exploratory analysis and for publication
# To learn more, follow this tutorial: http://www.rebeccabarter.com/blog/2017-11-17-ggplot2_tutorial/

# Assignment

# Resources
# ggplot2 cheat sheet: https://rstudio.com/wp-content/uploads/2015/03/ggplot2-cheatsheet.pdf

# The builtin dataset mpg gives fuel economy statistics for 38 popular models of cars

data("mpg")
mpg

# Your task is to produce visualizations that explore the data
#    Hint: You can use subset(mpg,factor=="value") to make a subset of the data
#       (for example, by manufacturer or year)

# Task 1: A scatter plot of city versus hwy, with different colors based on year (2008 or 1999). 
#    Make the size of the points related the number of cylinders the engine has
#    Hint: use factor(variable) to turn a number into a discrete variable



# Task 2: A boxplot or violin plot showing mpg (city or hwy) across each year
#     Bonus: Add jittered points to the plot with size related to the cylinder


# Task 3: Make a facet plot showing cty or hwy separated by the number of cylinders


# Identify a hypothesis you could test with statistical models that emerges from the plots you have made.



# Another exploration of data with identical mean, sd, and corr but with more fun patterns

group_by(datasaurus_dozen,dataset)
x.mean = tapply(datasaurus_dozen$x,datasaurus_dozen$dataset,mean)
y.mean = tapply(datasaurus_dozen$y,datasaurus_dozen$dataset,mean)
x.sd = tapply(datasaurus_dozen$x,datasaurus_dozen$dataset,sd)
y.sd = tapply(datasaurus_dozen$y,datasaurus_dozen$dataset,sd)

ggplot(datasaurus_dozen, aes(x=x, y=y, colour=dataset))+
  geom_point()+
  theme_void()+
  theme(legend.position = "none")+
  facet_wrap(~dataset, ncol=3)