Introduction

to
R Programming

Predictive Society and Data

Analytics Lab
Spring 2023
 Getting Started with R

• What is R?
• What can you do in R?
• Installing R & R Studio
Overview •
•
RStudio IDE
Help and Navigation
• Installing Packages

Programming in R

• Basic Operations
• Basic Object Types
• Data Structures
• Control structures
• User-defined Functions
• I/O
• Useful build-in Functions
• Plotting

Probability and Statistics in R

• Sampling
• Distributions
• Statistical Tests
FES - Predictive Society and Data Analytics Lab 2
Getting Started with R: What is R?
R Language
• A well-developed, simple and effective programming language based on S
• Supports imperative, object-oriented and functional programming paradigm
• A vast collection of libraries available for efficient data manipulation and analysis

R Environment
• Open-source and available for Ubuntu, MacOS and Windows
• Highly customizable, flexible and powerful with convenient integration with python, C++, etc.
• Effective data storage and handling facility for fast data manipulation and analysis
• Publish-quality graphics producible both on-screen and on hardcopy

FES - Predictive Society and Data Analytics Lab 3

Getting Started with R: What can you do in R?
• Simple arithmetic calculations • Data analysis
• Linear and nonlinear modelling, time series
>5+5 analysis, classification, clustering, etc
[1] 10 (caret, cluster, igraph, e1071, spacyr, Nnet)
• Advanced calculations • Statistical and Data manipulation packages
(stats, car, prob, stringr, dplyr, tidyr)
> 3*exp(2) + log(10)/sqrt(2)
• Data Visualization
[1] 23.79534
• Well-designed publication-quality plots using
ggplot2
• Object-oriented programming and
functional programming
• Most python libraries has interfaces or
packages for R

FES - Predictive Society and Data Analytics Lab | 4

Getting Started with R: Installing R & RStudio
R installation

• The latest version of R can be downloaded from the CRAN [Link]

• Choose the distribution of R suitable for your operating system

RStudio installation

• The latest version of RStudio can be downloaded from

[Link]

Using R with Anaconda or Jupyter

• Alternatively, you can also use R with Anaconda or Jupyter.

• More information found here
[Link]
FES - Predictive Society and Data Analytics Lab 5
Getting Started with R: R Studio IDE

Workspace

Source Scripts

Visualization/Help
Console

FES - Predictive Society and Data Analytics Lab 6

Getting Started with R: Help and Navigation
• The working directory
# set the working directory # Standard syntax
> setwd("~/folder") > help(FUN)
# get the current working directory # FUN = some function (e.g. 'sum')
> getwd() # Short-hand syntax
[1] "~/folder“ > ?FUN

• System functions > ??FUN

# example
# Check arguments of a function  ?sum
> args(FUN)
# Get help with functions
 help(FUN)

FES - Predictive Society and Data Analytics Lab 7

Getting Started with R: Installing Packages
• From CRAN • From Other Repos

# Installing packages from CRAN R # Installing packages from Bioconductor

repository
repository
> [Link](‘survival’)
> [Link]('BiocManager’)
> [Link](pkgs = ‘survival’)
> BiocManager::install(‘graph’)

# Load package
# Installing packages from GitHub
> library(survival)
> [Link](‘devtools’)
>
devtools::install_github(‘tidyverse/ggplot2
’)

8
 Getting Started with R

• What is R?
• What can you do in R?
• Installing R & R Studio
Overview •
•
RStudio IDE
Help and Navigation
• Installing Packages

Programming in R

• Basic Operations
• Basic Object Types
• Data Structures
• Control structures
• User-defined Functions
• I/O
• Useful build-in Functions
• Plotting

Probability and Statistics in R

• Sampling
• Distributions
• Statistical Tests
FES - Predictive Society and Data Analytics Lab 9
Programming in R: Basic Operations
•Assigning values to variables • Mathematical operators

# Using assignment operator '<-' ('=' * and / # Multiplication and

in other languages) division
> x <- (1 + 1/10%%3)^100 + and - # Addition and subtraction
> y <- 2^100 () # Parentheses: specify operation
> x == y order
[1] TRUE
# ^ or ** power:
# = is generally used for assigning 2^10 = 1024 = 2**10
values at function calls # %/% Integer division:
> pnorm(q, mean = 0, sd = 1, [Link] 10%/%3 = 3
= TRUE) # %*% Conformable matrix multi:
x %*% y
FES - Predictive Society and Data Analytics Lab | 10
Programming in R: Basic Object Types
• The basic elements are called “objects”

• Check type of an object

FES - Predictive Society and Data Analytics Lab 11

 Programming in R: String Example
# Defining a string # Splitting sentence into words # Regular Expression
message1 <- 'Hello Everyone' strsplit(message2, " ") message3<-c("New Hampshire", "New
print(message1) [[1]] Jersey","New Mexico", "New York")
[1] "Hello Everyone" [1] "Welcome" "to" "this" "Program"
message2 <- "Welcome to this Program" grep("Jersey", message3)
print(message2) # change string to uppercase [1] 2
[1] "Welcome to this Program" message_upper <- toupper(message1)
cat("Uppercase:", message_upper) grep("New",message3)
# double quote inside of double quotes [1] Uppercase: HELLO EVERYONE [1] 1 2 3 4
example1 <- "This is "R" programming"
Error: malformed raw string literal # change string to lowercase grepl("New",message3)
at line 1 message_lower <- tolower(message1) [1] TRUE TRUE TRUE TRUE
cat("\nLowercase:", message_lower)
# use of nchar() to find length of [1] Lowercase: hello everyone gsub('[ *]','_',message3)
message1 [1] "New_Hampshire"
nchar(message1) # Get part of the string "New_Jersey" "New_Mexico" "New_York"
[1] 14 substr(message2, 1, 7)
[1] "Welcome"
# use paste() to join two strings
paste(message1, ",", message2, sep = " ")
[1] "Hello Everyone , Welcome to this
FES - Predictive Society and Data Analytics Lab 12
Program”
Programming in R: Data Structures
• Data Structures • Environment is also a type of List
• Vectors (Same data type)
 Elements needs to have character
• Matrix (Same data type) names (named elements)
• Array (High-dimensional mix data)
• List (Mix data) • Indices start with 1 (not zero)
• Data Frame (Mix data)
• Several object functions available to
• Factor (Categorical Data)
apply a function to all data in a
structure (without looping)

 apply(), lapply(), sapply() etc.

FES - Predictive Society and Data Analytics Lab 13

Programming in R: Data Structures Examples
Vectors Matrices
a<-c(3, 4, 1, 7, 12) a<-seq(2, 16, 4) a<-matrix(c(1,2,3,4,5,6), nrow = 3, ncol = 2, byrow =
a a TRUE)
[1] 2 6 10 14 a
[1] 3 4 1 7 12 [,1] [,2]
names(a)<-LETTERS [1:4] [1,] 1 2
a[3] a [2,] 3 4
[1] 1 A B C D [3,] 5 6 rbind(c(2,5,3),c(4,3,4
2 6 10 14 ))
a[1,2] [,1] [,2] [,3]
a[c(2,5)] a[“C”] [1] 2 [1,] 2 5 3
[1] 4 12 C [2,] 4 3 4
10 a[1,]
a[2:4] [1] 1 2 a <- c(2,5,6,2)
[1] 4 1 7 dim(a) <- c(2,2)
a[c(1,3),] a
[,1] [,2] [,1] [,2]
a<-c(a,8) [1,] 1 2 [1,] 2 6
[1] 3 4 1 7 12 8 [2,] 5 6 [2,] 5 2

a<-append(a,10, after=length(a)) rownames(a) <- letters[1:3]

colnames(a) <- c("white", "black")
a
[1] 3 4 1 7 12 8 10 a[,”white”]
FES - Predictive Society and Data Analytics Lab 14
[1] 1 3 5
Programming in R: Data Structures Examples
Lists Arrays
a<-matrix(c(1,2,3,4), nrow = 2, ncol = 2, byrow = TRUE) a<-array(1:8, dim = (2,2,2))
b<<- list(a, “hello”, 3, c(4,2)) a[,,1]
b , , 1
[[1]] [,1] [,2]
[,1] [,2] [1,] 1 3
[1,] 1 2 [2,] 2 4
[2,] 3 4
a[2,2,1]
[[2]] [1] 4
[1] "hello"

[[3]]
[1] 3
Factor (Kind of a sorter)
[[4]] height <-factor(c(“small”, “medium”, “tall”,
[1] 4 2
“small”, “medium”))
names(b) <- LETTERS[1:4] height
b$C [1] small medium tall small medium
[1] 3 Levels: medium small tall

FES - Predictive Society and Data Analytics Lab 15

Programming in R: Data Structures Examples
# Create the data frame. # Get the structure of the data frame.
[Link] <- [Link]( str([Link])
emp_id = c (1:5), '[Link]': 5 obs. of 4 variables:
$ emp_id : int 1 2 3 4 5
emp_name = c("Rick","Dan","Michelle","Ryan","Gary" $ emp_name : chr "Rick" "Dan" "Michelle" "Ryan" ...
), $ salary : num 623 515 611 729 843
salary =c(623.3,515.2,611.0,729.0,843.25), $ start_date: Date, format: "2012-01-01" "2013-09-23" ...
start_date = [Link](c("2012-01-01", "2013-09-
23", "2014-11-15", "2014-05-11", "2015-03-27")), # Print the summary.
stringsAsFactors = FALSE summary([Link])
) emp_id emp_name salary
# Print the data frame. start_date
print([Link]) Min. :1 Length:5 Min. :515.2 Min. :2012-
01-01
emp_id emp_name salary start_date 1st Qu.:2 Class :character 1st Qu.:611.0 1st Qu.:2013-
1 1 Rick 623.30 2012-01-01 09-23
2 2 Dan 515.20 2013-09-23 Median :3 Mode :character Median :623.3 Median :2014-
3 3 Michelle 611.00 2014-11-15 05-11
4 4 Ryan 729.00 2014-05-11 Mean :3 Mean :664.4 Mean :2014-
5 5 Gary 843.25 2015-03-27 01-14
3rd Qu.:4 3rd Qu.:729.0 3rd Qu.:2014-
FES - Predictive Society and Data Analytics Lab 16
11-15
Programming in R: Data Structures Examples
# Extract Specific columns. # Add the "dept" coulmn.
result <-
[Link]$dept <-
[Link]([Link]$emp_name,[Link]$salar
y) c("IT","Operations","IT","HR","Finance")
print(result) v <- [Link]
[Link].emp_name [Link] head(v,2)
1 Rick 623.30 emp_id emp_name salary start_date
2 Dan 515.20 dept
3 Michelle 611.00 1 1 Rick 623.30 2012-01-01
4 Ryan 729.00 IT
5 Gary 843.25 2 2 Dan 515.20 2013-09-23
Operations
# Extract first two rows.
result <- [Link][1:2,] print(v)
print(result) emp_id emp_name salary start_date
emp_id emp_name salary start_date dept
1 1 Rick 623.3 2012-01-01 1 1 Rick 623.30 2012-01-01
2 - Predictive2Society and Data
Dan 515.2 2013-09-23 IT
FES Analytics Lab 17
2 2 Dan 515.20 2013-09-23
Programming in R: Data Structures Examples
# Merge two dataframes # Bind the two data frames by row
# Create a second data frame [Link] <- rbind([Link],[Link])
print([Link])
[Link] <- [Link](
emp_id = c (6:8), emp_id emp_name salary start_date dept
emp_name = 1 1 Rick 623.30 2012-01-01
c("Rasmi","Pranab","Tusar"), IT
salary = c(578.0,722.5,632.8), 2 2 Dan 515.20 2013-09-23
start_date = [Link](c("2013-05-21", Operations
"2013-07-30","2014-06-17")), 3 3 Michelle 611.00 2014-11-15
dept = IT
c("IT","Operations","Fianance"), 4 4 Ryan 729.00 2014-05-11
stringsAsFactors = FALSE HR
) 5 5 Gary 843.25 2015-03-27
Finance
6 6 Rasmi 578.00 2013-05-21
IT
FES - Predictive Society and Data Analytics Lab 18
7 7 Pranab 722.50 2013-07-30
Programming in R: Control Structures
• If-conditions • For loop
 First  3.  First

 d

 Second

 Second

 third
FES - Predictive Society and Data Analytics Lab 19
Programming in R: Control Structures
While loop • Other Functions usable for control
1. Ifelse() for vectors

2. Switch() to replace nested ifs

Repeat Loop

FES - Predictive Society and Data Analytics Lab 20

Programming in R: User-defined Functions
• Two-ways to define a function • Scope rules applies to R Functions
• The following will cause error

• Returning values

• Call with arguments or named arguments

• Can have many inputs but one

output (return data structure if many outs)
FES - Predictive Society and Data Analytics Lab 21
Programming in R: I/O
Reading
To load specified data sets, or list available in an R session use data().
from > library(survival) # Load datasets in survival
package
Packages > data(cancer)

Reading
To load files saved using R save() (i.e., ‘.rda’ or ‘.Rdata’) use load()
Data saved # Load .rda or .Rdata files
> load(“[Link]”)
by R
[Link]() # read a comma seperated file [Link]() # read a file in table format
[Link]() # read a TAB delimited file [Link]() # read a fixed width
From other format table
file types [Link]() # read data interchange format file
file
readLines() # Read individual lines from a

scan() # foundation for most other read functions

FES - Predictive Society and Data Analytics Lab 22

Programming in R: I/O
save(a,file="filename")

save(a, b,c,d,file="filename")
Saving
e<-c("a","b","c","d") save(list=e,file="filename")
Data in R
save(list=ls(all),file="filename")

[Link](file="filename")

Save to [Link]() #read a comma seperated file

other types write.csv2() #read a file with TAB
[Link]() #read a file in table format
of files

FES - Predictive Society and Data Analytics Lab 23

Programming in R: Useful Built-in Functions
• Mathematical Functions
• atan(), sin(), cos(), tan(), exp(), log(), log10(), sqrt(), floor(),
ceiling(), round(), abs(), factorial(), trunc()

• Vector Functions
• ave(), sort(), dim(), sum(), prod(), sample(), mean(), median(),
union(), unique(), sd(), quantilie(), range(), diff(), min(), max()

• String Functions
• substr(), grep(), sub(), paste(), strsplit(), tolower(), toupper()

FES - Predictive Society and Data Analytics Lab | 24

Programming in R: Plotting
• Plots are used for data visualization in R, use “? • Plotting with ggplot2 Object
plot” to see the full description
• Powerful plotting package with a wide variety
# Examples of options
plot(readData$Height, type="l",
col=”black")
[Link]("ggplot2")
boxplot(myDF, col = rainbow(3))
library(ggplot2)
hist(readData$Weight)
• Saving plots in different file format ?ggplot2
# Assign type to give
pdf(”[Link]", width=5, height=6) ggplot([Link](x = c(-3, 3)), aes(x = x))
jpeg(”[Link]", width=5, height=6) + stat_function(fun = pnorm)
png(”[Link]", width=5, height=6)
# plot and save
plot(readData$Height, col="blue”)
[Link]()

FES - Predictive Society and Data Analytics Lab 25

Programming in R: Plotting Example

FES - Predictive Society and Data Analytics Lab The data taken from 26
[Link]
Programming in R: Plotting Example

FES - Predictive Society and Data Analytics Lab The data taken from 27
[Link]
Programming in R: Plotting Example
• Built-in color options in R

FES - Predictive Society and Data Analytics Lab 28

Programming in R: Plotting Example

FES - Predictive Society and Data Analytics Lab 29

Programming in R: Plotting Example

FES - Predictive Society and Data Analytics Lab The data taken from 30
[Link]
Programming in R: Plotting Example

FES - Predictive Society and Data Analytics Lab The data taken from 31
[Link]
Programming in R: Plotting Example

FES - Predictive Society and Data Analytics Lab The data taken from 32
[Link]
Programming in R: Plotting Example

FES - Predictive Society and Data Analytics Lab The data taken from 33
[Link]
 Getting Started with R

• What is R?
• What can you do in R?
• Installing R & R Studio
Overview •
•
RStudio IDE
Help and Navigation
• Installing Packages

Programming in R

• Basic Operations
• Basic Object Types
• Data Structures
• Control structures
• User-defined Functions
• I/O
• Useful build-in Functions
• Plotting

Probability and Statistics in R

• Sampling
• Distributions
• Statistical Tests
FES - Predictive Society and Data Analytics Lab 34
Probability and Statistics in R: Sampling
• Sampling from normal distribution
• Sampling from distribution
example
# Sampling without replacement
R <- 10
D <- rnorm(50) # Random normal
• We can use the following distribution
sampled <- sample(D, R) # By default replace=F
functions to generate a random
sample from common # Sampling with replacement
distributions such as normal, chi- R <- 10
D <- rnorm(50, 5, 2) # Random normal
squared, poisson, etc
distribution with mean = 5 and sd = 2
sampled <- sample(D, R, replace=T)
sampled
• Use “?rnorm” for details
FES - Predictive Society and Data Analytics Lab 35
Probability and Statistics in R: Distributions
• Each Distribution has functions with
p, q, d and r for the following
• p for "probability", the cumulative distribution
function (c. d. f.)
• q for "quantile", the inverse c. d. f.
• d for "density", the density function (p. f. or
p. d. f.)
• r for "random", a random variable having the
specified distribution

# Most common random distribution functions

rnorm, runif, rpois, rmvnorm, rnbinom, rbinom, rbeta,
rchisq, rexp, rgamma, rlogis, rstab, rt, rgeom, rhyper,
rwilcox, rweibull
FES - Predictive Society and Data Analytics Lab 36
Probability and Statistics in R: Table Lookup
• Looking up Normal and Standard • Looking up Chi-Square and Student-
Normal Curves T Distributions
# Normal distribution look up for any mean and # Chi-square look up
sd qchisq(.95, df=7)
dnorm(27.4, mean=50,sd=20) # P(X=27.4) pf
pnorm(27.4, mean=50, sd=20) # P(X<27.4) cdf # Student-t Lookup
qnorm(0.95, mean=100, sd=15) # inverse look up qt(.975, df=5)

# Standard Normal look up, same function

without mean, sd
dnorm(x)
pnorm(x)
qnorm(p)

FES - Predictive Society and Data Analytics Lab 37

Probability and Statistics in R: Statistical Tests
# Student´s T-Test
[Link](x, y = NULL, alternative = c("[Link]", "less", "greater"), mu = 0, paired = FALSE, …)
# Wilcoxon Test
[Link](x, y = NULL, alternative = c("[Link]", "less", "greater"), mu = 0, paired = FALSE, exact =
NULL, …)
# Shapiro-Wilk Test
[Link](x)
# Kolmogorov-Smirnov Test
[Link](x, y, …, alternative = c("[Link]", "less", "greater"), exact = NULL, …)
# Pearson's Chi-squared Test
[Link](x, y = NULL, correct = TRUE, p = rep(1/length(x), length(x)))
# Correlation Test
[Link](x, y, alternative = c("[Link]", "less", "greater"), method = c("pearson", "kendall",
"spearman"), exact = NULL, [Link] = 0.95, continuity = FALSE, …)

FES - Predictive Society and Data Analytics Lab 38

Probability and Statistics in R: Example

FES - Predictive Society and Data Analytics Lab 39

Final Notes
• Learn more on R by referring to
Mathematical Foundations of Data
Science Using R
• Available in TAU library
• For question regarding the video
• [Link]@[Link]
• The slides are based on content from
book and Aliyu Musa's Tutorial on R

FES - Predictive Society and Data Analytics Lab 40