Announcement

Changes in Office Hours

• Office: South Hall 117B.
• Office Hours:
  • Mondays 3:30 PM-5:30 PM
  • Wednesdays 1:30 PM-3:30 PM.

Installing the Tools

R programming

The R language is available as a free download from the R Project website at:

• Windows: https://cran.r-project.org/bin/windows/base/
• Mac: https://cran.r-project.org/bin/macosx/
  • Download the file of R that corresponds to your Mac OS (Big Sur, Apple silicon arm64, High Sierra, El Capitan, Mavericks, etc.)

Installing the Tools

RStudio

• RStudio offers a graphical interface to assist in creating R code:

  • The RStudio Desktop is available as a free download from the following webpage:
  • https://www.rstudio.com/products/rstudio/download/#download

Installing the Tools

RStudio Environment

Installing the Tools

RStudio Environment

Installing the Tools

RStudio Environment

Installing the Tools

RStudio Environment

Installing the Tools

RStudio Environment

# Answer "no" to:
# Do you want to install from sources the packages which need compilation?
update.packages(ask = FALSE, checkBuilt = TRUE)
pkgs <- c("tidyverse", "nycflights13", "gapminder", "skimr")
install.packages(pkgs,
    dependencies = c("Depends", "Imports", "LinkingTo"))Copy Code

Management of Files, Directories, and Scripts

Materials for the book, Practical Data Science with R

• Click the green "Code" button and download the ZIP file from the following GitHub page: https://github.com/WinVector/PDSwR2.

Management of Files and Directories

Finding the path name of the file

Management of Files, Directories, and Scripts

Code and comment style

• The two main principles for coding and managing data are:

  • Make things easier for your future self.
  • Don't trust your future self.

• The # mark is R's comment character.

  • # indicates that the rest of the line is to be ignored.
  • Write comments before the line that you want the comment to apply to.

• Consider using block commenting for separating code sections.

  • #### defines a coding block.

• Break down long lines and long algebraic expressions.

Starting with R

RStudio Options Setting

Starting with R

• Let's try a few commands to help you become familiar with R and its basic data types.

1
1/2
'Joe'
"Joe"
"Joe"=='Joe'
c()
is.null(c())
is.null(5)Copy Code

c(1)
c(1, 2)
c("Apple", 'Orange')
length(c(1, 2))
vec <- c(1, 2)
vecCopy Code

Starting with R

Code and comment style

• R has many assignment operators (e.g., <-, =, -> ).
• The preferred one is <-.

x <- 2
x < - 3
print(x)
x <- 5
x = 5
5 -> xCopy Code

Starting with R

Shortcuts

Starting with R

R data types

• Primary data types in R are as follows:
  • Logical: A simple binary variable that may have only two values---TRUE or FALSE.
  • Numeric: Decimal numbers
  • Integer: Integers
  • Character: Text strings
  • Factor: Categorical values. Each possible value of a factor is known as a level.
  • Ordered Factor: A special factor data type where the order of the levels is significant. E.g., Low, Medium, and High

Starting with R

R data types

• Test the data types.

  x <- TRUE
  y <- 1
  z <- 'Data Analytics'
  productCategory <- c('fruit', 'vegetable', 'dry goods', 'fruit',
                   'vegetable', 'dry goods')
  productCategoryFactor <- factor(productCategory)Copy Code

• The class() function returns the data type of an object.
  • What are classes for x, y, z, productCategory, and productCategoryFactor?

Starting with R

R data types

• Most R data types are mutable, in that we're allowed to change them.

a <- c(1, 2)
b <- a
print(b)
# Alters a
a[[1]] <- 5
print(a)
print(b)Copy Code

Starting with R

Lists

• Lists, unlike vectors, can store more than one type of object.
  • The ways to access items in lists are the $ operator and the [[]] operator.

x <- list('a' = 6, b = 'fred')
names(x)
x$a
x$b
x[['a']]
x[c('a', 'a', 'b', 'b')]Copy Code

Starting with R

R data types

• Here are examples of a vector and a list.

example_vector <- c(10, 20, 30)
example_list <- list(a = 10, b = 20, c = 30)
example_vector[1]
example_list[1]
example_vector[[2]]
example_list[[2]]
example_vector[c(FALSE, TRUE, TRUE)]
example_list[c(FALSE, TRUE, TRUE)]
example_list$b
example_list[["b"]]Copy Code

Starting with R

Errors

• Errors are just R's way of saying it safely refused to complete an ill-formed operation

• Fear of errors should not limit experiments.

x <- 1:5
print(x)
x <- meanMISSPELLED(x)  
print(x)                                 
x <- mean(x)                      
print(x)Copy Code

Starting with R

Data Frames

• R’s central data structure is the data frame.
• A data frame is organized into rows and columns.
• Data frames are essentially lists of columns.
• Data frames can have columns of different types.

d <- data.frame(x=c(1,2),
              y=c('a','b'))
d[['x']]
d$x
d[[1]]Copy Code

d
d[1,]
d[,1]
d[1,1]
d[1, 'x']Copy Code

Starting with R

Data Frames

• The R data.frame class is designed to store data in a very good "ready for analysis" format.

d <- data.frame(col1 = c(1, 2, 3), col2 = c(-1, 0, 1))
print(d)
d$col3 <- d$col1 + d$col2
print(d)Copy Code

Starting with R

NULL and NA values

• NULL is just an alias for c(), the empty vector.
• NA indicates missing or unavailable data.

c(c(), 1, NULL)
c("a", NA, "c")Copy Code

Starting with R

NULL and NA values

• NULL is just an alias for c(), the empty vector.
• NA indicates missing or unavailable data.

c(c(), 1, NULL)
c("a", NA, "c")Copy Code

Starting with R

NULL and NA values

• Most R data types are mutable, in that we're allowed to change them.

d <- data.frame(x = 1, y = 2)
d2 <- d
d$x <- 5
print(d)
print(d2)Copy Code

Working with Data from Files

• Step 1. Find the path name for the file, car.data.csv, from the sub-folder, 'UCICar' in the folder, 'PDSwR2-main'.

• Step 2. In the code below, replace 'PATH_NAME_FOR_THE_FILE_car.data.csv' with the path name for the file, car.data.csv.

• Step 3. Run the following R code:

uciCar <- read.table(
        'PATH_NAME_FOR_THE_FILE_car.data.csv',
        sep = ',',
        header = TRUE,
        stringsAsFactor = TRUE )
View(uciCar)Copy Code

Working with Data from Files

Examining data frame

• class() tells you what kind of R object you have.
• dim() shows how many rows and columns are in the data for data.frame.
• head() shows the top few rows of the data.
• help() provides the documentation for a class.
  • Try help(class(uciCar)).
• str() gives us the structure for an object.

Working with Data from Files

Examining data frame

• summary() provides a summary of almost any R object.
• skimr::skim() provides a more detailed summary.
• print() prints all the data.
  • Note: for large datasets, this can take a very long time and is something you want to avoid.
• View() displays the data in a simple spreadsheet-like grid viewer.
• dplyr::glimpse() displays brief information about the data.

Working with Data from Files

Examining data frame

print(uciCar)
class(uciCar)
dim(uciCar)
head(uciCar)
help(class(uciCar))
str(uciCar)
summary(uciCar)
library(skimr)
skim(uciCar)
library(tidyverse)
glimpse(uciCar)Copy Code

Working with Data from Files

Reading from an URL

• We can import the data file from the web.

tvshows <- read.table(
        'https://bcdanl.github.io/data/tvshows.csv',
        sep = ',',
        header = TRUE,
        stringsAsFactor = TRUE)Copy Code

Working with Data from Files

ggplot

• Let's try some data visualization using ggplot():

ggplot(tvshows) + 
  geom_point(aes(x=GRP, y=PE, color=Genre))
ggplot(tvshows) + 
  geom_point(aes(x=GRP, y=PE)) + 
  facet_wrap(~Genre)Copy Code

• What is the nature of the the relationship between audience size (GRP) and audience engagement (PE)?