R Notes

Introduction

R can be regarded as an implementation of the S language which was developed at Bell Laboratories
Expression-language, case sensitive
Special symbols allowed are . (dot) and _ (underscore)
Variable Name must start with ‘.’ or a letter, and if it starts with ‘.’ the second character must not be a digit. Names are effectively unlimited in length.
Semicolon or newline separate commands
Execute commands stored in a R file
```
source("commands.R")
```

Output data to an external file

sink("record.lis") # writes the output to record.lis instead of printing it to stdout

objects() or ls() returns the object currently stored in the session

rm() can be used to remove an object

rm(list = ls()) # remove every object present in memory

Packages

library() function (with no args) returns list of installed packages
getOption("defaultPackages") returns the list of packages automatically loaded during startup
chooseCRANmirror() function provides a menu to choose a mirror
setRepositories() function provides a menu to select the repos you want to use
update.packages() simply updates all existing packages
search() returns list of loaded packages

install.packages("pacman") # install the package manager
# install.packages(c("pkg1", "pkg2")) # install multiple packages at once

library(pacman) # import without a conformation message, not known
require(pacman) # import with a confirmation message

pacman::p_load(pacman, ggplot2) # alternative way of loading packages

# detach(ggplot2, unload=TRUE) # detach a package, and unload the namespace
p_unload(ggplot2) # unload package/s
p_unload(all) # unload all packages

# remove.packages(c("pkg1", "pkg2")) # remove multiple packages at once

Assignment

x <- c(5, 4, 3, 2, 1)
# c(5, 4, 3, 2, 1) -> x # alternative
y <- c(1, 2, 3, 4, 5)
assign("z", c(10, 20, 30))

Data Types

String

paste("hello, ", name, sep="") # concatinate multiple values
# instead of setting sep="", you can just use paste0()

Character
Integer
Double

Array

lt <- 1:10 # create a list 'lt' with number from 1 to 10
abs(lt) # applies abs() to every element of lt and prints it

lt * 2 # returns a new vector with each element multiplied by 2

c(lt, 0, lt) -> js # stores a vector consisiting of the elements of lt, then 0 and again the elements of lt

Type conversion

as.integer(), as.double(), as.character()

x <- "10" # 10 is a string here
as.integer(x) # returns x as integer

Special

Inf and -Inf represent infinite values
NA represent Not Available
NaN means Not a Number
NULL means nothing

Input/Output

Read a value from user
```
readline("What is your name?")
```

Print value to stdout

print("hello") # directly print values to console

File Handling

Create a new file
```
file.create("hello.R")
```
Change working directory
```
setwd("./code")
```

Read table from a file

read.table("votes.csv")
# votes <- read.table("election.csv", sep=",", header = TRUE) # read a csv file

Read table from a .csv file
```
read.csv("votes.csv")
```

Save a R object into a .Rdata file

save(dailyCosts, file = "dailyCosts.Rdata")

Load a R object from file

load("dailyCosts.Rdata") # creates a variable dailyCosts containing the data of loaded file

Objects

View(obj_name) opens another tab to display an object

Tables (Frames)

tbName[row, column] gives a element stored at particular row and column of a table
tbName[, 1] gives the first column of table
tbName$columnName gives you list of elements in the given column
rowNames(votes) gives you the list of row names, and colNames(votes) gives the list of column names
ncol(votes) returns number of columns and nrow(votes) returns number of columns
unique(votes$voter_category) returns a vector of unique values present inside voter_category column

votes <- read.csv("election.csv")
# votes$poll # returns vector of values in poll column
# votes$poll + votes $mail # this adds up the values of the given two columns for each row
votes$total <- votes$poll + votes$mail # create a new column that contains sum of poll and mail value for each row

write.csv(votes, "election.csv", row.names = FALSE) # write data back to a csv file, row.names prints the row indexes

factor(votes$Q21,
       label = c("Yes", "No", "Undecided"),
       exclude = c(-1)
       ) # replaces the values with labels, and excludes -1

votes <- votes[!is.na(votes$poll)] # remove rows where poll is NA

subset() method can be used to generate subsets

# generate a subset of spending table, where the Item Name column is equal to "Lunch"
milkSpend <- subset(spending, `Item Name` == "Lunch")

rbind() can be used to combine multiple data frames

sales <- rbind(q1, q2, q3, q4) # combine sales from all four quarters

Vectors

Create a vector from existing vector (based on positions)

marks <- c(15, 19, 10, 9, 20, 5)
marksFailed <- marks[c(4, 6)] # 9, 5
marksPassed <- marks[-c(4, 6)] # 15, 19, 10, 20

Relational operations on vectors

marks <- c(15, 19, 10, 9, 20, 5)
marks > 10 # TRUE TRUE FALSE FALSE TRUE FALSE

which() function returns the indices of elements that are true

marks <- c(15, 19, 10, 9, 20, 5)
which(marks > 10) # 1 2 5
marks[(which(marks > 10))] # 15, 19, 20

vectors use & and | for logical operations, while && and || are used for single value variables
any() function tells if there is any value that satisfies the condition or not
```
marks <- c(15, 19, 10, 9, 20, 5)
any(marks > 10)
```
all() is the opposite of any() and tells if all of the elements of vector satisfy the condition or not

Functions

x <- -12;
abs(x); # 12

data(); # returns info about built in datasets

head(mtcars); # shows first 6 entries of mtcars dataset

str(mtcars); # display the structure (column name, its data type and initial values) of mtcars dataset

sum(x, 12, 30)

Defining functions

getInt <- function(prompt = "Enter Integer: ") {
    votes <- as.integer(readline(prompt))
    # return(votes) # ommited as R automatically returns last value
}

Data Visualization

R by default provides graphics library to visualize data, other more advanced packages can also be used like ggplot2
Line Types (lty)
1. Solid
2. Dashed
3. Dotted
4. DotDash
5. LongDash
6. TwoDash

Scatter Plot

plot (
    x = mtcars$mpg,
    y = mtcars$wt,
    main = "Scatter Plot: Weight vs Miles Per Gallon",
    xlab = "Miles Per Gallon",
    ylab = "Weight",
    col = "red",
    pch = "*", # plotting character (can be a symbol or a integer code)
    cex = 1.5 # represents the size of plotting character (1 means default)
)

grid() # turns on grid

Single Line Plot

time_vec  <- 1:10
var1      <- cumsum(rnorm(10))

plot(
  x = time_vec,
  y = var1,
  type = "b",
  col = "darkgreen",
  lwd = 2, # line width
  pch = 16,
  main = "Single Line Plot",
  xlab = "Day",
  ylab = "Value"
)

Matplot (multiple lines on one set of axes)

time_vec  <- 1:10
var1      <- cumsum(rnorm(10))
var2      <- cumsum(rnorm(10))

mat_bind <- cbind(var1, var2)

matplot(
  x = time_vec,
  y = mat_bind,
  type = "b",
  pch = 1, # point character
  col = c("darkgreen", "darkblue"),
  lty = 1, # line type
  lwd = 2, # line width
  main = "Matplot / Multi-line plot",
  xlab = "Day",
  ylab = "Values",
  # ylim=c(1, 40), # pass a range vector to specify limit
)

legend("topleft", legend = c("Series A", "Series B"), col = c("darkblue", "darkgreen"), pch = 1, lty = 1)

Bar Plot

cat_data  <- table(diamonds_sample <- sample(c("Ideal", "Premium", "Good"), 100, replace = TRUE))

barplot(
    cat_data, # height parameter
    main = "Bar Plot: Cut Frequencies",
    xlab = "Cut Quality",
    ylab = "Count",
    col = c("coral", "steelblue", "gold"), # different color for each of three bars
    border = NA, # no borders
    ylim = c(0, 40)
    )

Pie Plot

cat_data  <- table(diamonds_sample <- sample(c("Ideal", "Premium", "Good"), 100, replace = TRUE))

pie(
  x = cat_data,
  main = "Pie Plot: Quality Distribution",
  col = c("coral", "steelblue", "orange"),
)

Histogram

num_data  <- rnorm(150, mean = 50, sd = 10)
hist(
    num_data,
    main = "Histogram of Values",
    xlab = "Measured Value",
    col = "lightblue",
    border = "white",
    )

Density Plot

d <- density(num_data)
plot(
    d,
    main = "Density Plot",
    xlab = "Value",
    col = "darkred",
    lwd = 2
)
polygon(d, col = rgb(0.8, 0.2, 0.2, 0.3), border = "darkred") # Fills area under curve

Dot Chart

dotchart(
  mtcars$mpg,
  labels = rownames(mtcars),
  cex = 0.7,
  main = "Dot Plot: Miles Per Gallon",
  xlab = "MPG",
  color = "darkgreen"
)

Venn Diagram

# Sample set sizes
setA <- 1:100
setB <- 50:150
overlap <- length(intersect(setA, setB))

# Plot blank canvas (plain white canvas)
plot(0, 0, type = "n", xlim = c(-3, 3), ylim = c(-3, 3), axes = FALSE, xlab = "", ylab = "", main = "Base R Venn Concept")

# Draw two overlapping circles using symbols()
symbols(x = c(-0.6, 0.6), y = c(0, 0), circles = c(1.2, 1.2), add = TRUE, inches = FALSE, bg = c(rgb(1,0,0,0.3), rgb(0,0,1,0.3)))
# circles diameter is 1.2, and are half immersed into each other (that is 0.6)

# Add set labels
text(x = c(-1.2, 0, 1.2), y = 0, labels = c(length(setA) - overlap, overlap, length(setB) - overlap), font = 2)
# -1.2, 0, 1.2 are three points representing the positions of three numeric values that present sets cardinality

# Add legend
legend("topleft", legend = c("Set A", "Set B"), fill = c(rgb(1,0,0,0.5), rgb(0,0,1,0.5)))

Help

help.start() loads the R documentation in html
If you use help() it will look for the item’s help page. If it doesn’t encounter any conflicts, then it will just present the man page otherwise it will show the search results
Alternatively you can use help.search("crayon") or ??crayon ¹ to search for man pages, instead of directly loading the relevant one
example("crayon") ¹ will return an example usage of the topic

?mpg # returns info about mpg dataset
help(mpg) # alternative way

?View() # returns info about View() function

FAQ

Assignment in R: Assignment operator or equals sign?

Rstudio

shortcuts

C-RETURN - Run the code (runs the current line in console)
C-l - clear console

References

R Manual

Crayon is a package for colored terminal output ↩︎ ↩︎

Introduction#

Packages#

Assignment#

Data Types#

Type conversion#

Special#

Input/Output#

File Handling#

Objects#

Tables (Frames)#

Vectors#

Functions#

Data Visualization#

Scatter Plot#

Single Line Plot#

Matplot (multiple lines on one set of axes)#

Bar Plot#

Pie Plot#

Histogram#

Density Plot#

Dot Chart#

Venn Diagram#

Help#

FAQ#

Rstudio#

shortcuts#

References#