Introduction to R

• Open source
• Has, arguably, the best graphics engine of statistical software packages
• Only does what you ask
• Widely used in industry, academia, and government

On your machine

• Go to http://cran.r-project.org/mirrors.html
• Pick a download site (0-Cloud is a good choice)
• Download the base package for your operating system:
  • Download R 3.1.3 (current version)
  • Run the installation script
• Go to https://www.rstudio.com/ide/download/ and download Rstudio

On the server

• Go to http://rstudio.lawrence.edu/
• Log in using your LU credentials
• Note: Rstudio server can only be accessed on campus

In console

In editor

• Treat learning R like learning a new language, it takes a while to learn the

  • grammar
  • syntax
  • vocabulary

• You don't need to memorize all of the commands, make your life easier by using

  • help files
  • cheat sheets
  • Google

The console can be used as a calculator

1+1
15.3 * 23.4 
5^5

There are many built-in math functions available

pi
cos(pi)
exp(1)
abs(-6) 
factorial(4) 
sqrt(16) 

You can store results as named variables for later use

product <- 15.3 * 23.4 # save result
product                # display the result

[1] 358.02

Now you can use the variable product

0.5 * product

[1] 179.01

log(product) # natural log

[1] 5.880589

If you have very few observations you can enter them by hand for each variable as a vector

a <- c(3, 5, 9, 7, 8, 1)
a

[1] 3 5 9 7 8 1

b <- 1:6
b

[1] 1 2 3 4 5 6

For plotting and modeling, we will want to deal with data.frame objects

df <- data.frame(a = a, b = b)
df

  a b
1 3 1
2 5 2
3 9 3
4 7 4
5 8 5
6 1 6

• If you already have a data set saved, then you can simply load the data set into R.

• Example: If you wanted to read in the tips.csv data set, then we can run the command

    tips <- read.table(file = file.choose(), sep = ",", header = TRUE)
  

  • go find the tips.csv file once you get the pop-up window

• read.table is our workhorse function, and can read in numerous file types

• for different file types you will need to specify different field separator characters:

  Separator	Description
  sep = " "	white space separated
  sep = "\t"	tab separated
  sep = ","	comma separated files (.csv)

• Use header = TRUE if there are column names

Try out the following commands

dim(tips)
str(tips)
head(tips)
summary(tips)

We can access columns of the data set using a command of the following form

data$col.name

or by using the column number

data$[,col.num]

Review

• We will use the ggplot2 package for graphics

• If you are using your personal computer, you will need to install this package before you use it the first time

  install.packages("ggplot2")
  

• You will need to load this package at the beginning of each R session:

  library(ggplot2)
  

With only one variable, qplot guesses that you want a bar chart or a histogram

qplot(tip, data = tips)
qplot(day, data = tips)

To obtain a boxplot use the following command

qplot(x = factor(0), y = tip, data = tips, geom = "boxplot") + 
  xlab("") +
  scale_x_discrete(breaks = NULL) + 
  coord_flip()

Change the bin width of histograms

qplot(tip, data = tips, binwidth = 1)
qplot(tip, data = tips, binwidth = 0.5)
qplot(tip, data = tips, binwidth = 0.01)

Adding other variables

You can use aesthetics or faceting to add additional variables to your plots.

qplot(tip, data = tips, binwidth = .5)
qplot(tip, data = tips, binwidth = .5, fill = sex)
qplot(tip, data = tips, binwidth = .5, fill = factor(size))

qplot(tip, data = tips, binwidth = .5) + 
  facet_wrap(~size)

Review

qplot(x = totbill, y = tip, data = tips, geom = "point")

qplot(x = totbill, y = tip, data=tips, geom = "point") + geom_smooth(method = "lm")

Let's look at tipping percentage now. First, we'll have to define the new variable

tips$tip.pct <- 100 * tips$tip / tips$totbill

Now, let's look at pairs of variables

qplot(x = sex, y = tip.pct, geom = "boxplot", data = tips)

qplot(x = smoker, y = tip.pct, geom = "boxplot", data = tips)

qplot(x = sex:smoker, y = tip.pct, geom = "boxplot", data = tips)