Note_Tech

All technological notes.

Project maintained by simonangel-fong Hosted on GitHub Pages — Theme by mattgraham

R - Fun

Back

R - Fun

Features
- Software that provides programming environment for Statistical Data Analysis
- Based on S
- Interpreted Language
- Data Storage, Analysis, Graphing
- Free and Open-Source Software
Strengths
- Free and Open Source
- Strong User Community
- Highly extensible, flexible
- Implementation of high-end statistical methods
- Flexible graphics and intelligent defaults
Weakness
- Steep learning curve
- Slow for large datasets
Fundamental
- Highly Functional
  - Everything done through functions
  - Strict named arguments
  - Abbreviations in arguments OK (e.g. T for TRUE)
- Object Oriented
  - Everything is an object
  - “<-” is an assignment operator
  - “X <- 5”: X GETS the value 5
Programming in R
- Functions & Operators typically work on entire vectors
- Expressions surrounded by {}
- Codes separated by newlines, “;” not necessary
- You can write your own functions and use them
Descriptive Statistics
- Has functions for all common statistics
- summary() gives lowest, mean, median, first, third quartiles, highest for numeric variables
- stem() gives stem-leaf plots
- table() gives tabulation of categorical variables
Statistical Modeling
- Numerous libraries & packages
- Distinction between factors and regressors
  - factors: categorical, regressors: continuous
  - you must specify factors unless they are obvious to R
  - dummy variables for factors created automatically
- Use of data.frame makes life easy

Help in R

From Documentation:
- ?WhatIWantToKnow
- help(“WhatIWantToKnow”)
- help.search(“WhatIWantToKnow”)
- help.start()
- getAnywhere(“WhatIWantToKnow”)
- example(“WhatIWantToKnow”)
Documents: “Introduction to R”
Active Mailing List
- Archives
- Directly Asking Questions on the List

Basic Math in R

When you input a mathematical expression in the R console, it functions as a calculator
Common arithmetic operations and mathematical functionality are available at the console prompt
Addition, subtraction, multiplication and division can be performed with the +, -, * and / symbols respectively
Exponents (powers or indices) can be performed using ^.
The order of calculations in a single command can be controlled with parentheses ()
Square root, logarithms and exponents can be performed with the sqrt(), log() and exp() functions respectively

Assigning Objects

In R, assignment can be specified in two ways:
- Using arrow notation (<-)
- Using a single equal sign (=).

Data Structures in R

Supports virtually any type of data
Numbers, characters, logicals (TRUE/ FALSE)
Arrays of virtually unlimited sizes
Simplest: Vectors and matrices
Lists: Can contain mixed type variables
Data Frame: Rectangular/Tabular dataset

	Linear	Rectangular
All Same Type	Vectors	Matrix
Mixed	List	Data Frame

built-in data types
- Atomic classes: numeric, logical, character, integer, complex vectors, lists
- Factors
- Data frames and matrices

Vectors

Essential building block for handling multiple items in R.
Homogeneous data structures.
Can be numeric, character, logical, etc.
Created using c() function.
Example:

my.vector <- c(1.5, 2.9, 3.0)
my.vector[5:8]

len(my.vector)
class(my.vector)
sort(my.vector, decreasing = TRUE)
rep(x=10, times = 5) ## Repeat 10, 5 times
seq(from=3,to=27,by=3)
## Returns a vector of values from 3 to 27 with interval of 3.

# Explicit Coercion
my.vector <- c(1.5, 2.9, 3.0)
as.numeric(my.vector) ## Numeric
as.character(my.vector) ## Character

# Implicit Coercion
my.vector <- c(1.5, "a") ## Character
my.vector <- c(2, TRUE) ## Numeric
my.vector <- c(TRUE, "a") ## Character

Lists

Lists are highly valuable data structures in R
Heterogeneous data structures
Can contain elements of different data types
A single list can contain diverse elements such as a numeric matrix, a logical array, a character string, and a factor object.
Created using list() function

my.list <- list(matrix(data=1:4,nrow=2,ncol=2), c(F,F,T,T),"Hello World")
# [[1]]
#      [,1] [,2]
# [1,]    1    3
# [2,]    2    4

# [[2]]
# [1] FALSE FALSE  TRUE  TRUE

# [[3]]
# [1] "Hello World"

my.list <- vector("list", length = 5) ## Create an empty list of a pre-specified length
# [[1]]
# NULL

# [[2]]
# NULL

# [[3]]
# NULL

# [[4]]
# NULL

# [[5]]
# NULL
my.list[[1]] ## Read data from index 1. Indexes are entered in double square brackets.
# NULL

length(my.list) ## To find out the length of a list
# 5

Matrix

Matrix is simply several vectors stored together
Two-dimensional data structures
Elements are of the same data type
The size of a matrix is specified by number of rows and number of columns.
Created using matrix() function
Example:

my.matrix <- matrix(c(1,2,3,4,5,6,7,8,9), nrow = 3, ncol = 3)
my.matrix <- matrix(c(1,2,3,4,5,6,7,8,9), nrow = 3, ncol = 3, byrow = FALSE)
## It will fill data column-wise fashion 先沿着col, 再row
#      [,1] [,2] [,3]
# [1,]    1    4    7
# [2,]    2    5    8
# [3,]    3    6    9
my.matrix <- matrix(c(1,2,3,4,5,6,7,8,9), nrow = 3, ncol = 3, byrow = TRUE)
## It will fill data row-wise fashion. 先沿着row, 再col
#      [,1] [,2] [,3]
# [1,]    1    2    3
# [2,]    4    5    6
# [3,]    7    8    9

dim(my.matrix) ## Returns the size of a matrix
# [1] 3 3

# Vector to matrix directly by adding dimension attribute
my.vector <- c(1,2,3,4,5,6,7,8,9)
dim(my.vector) <- c(3,3) ##
#      [,1] [,2] [,3]
# [1,]    1    4    7
# [2,]    2    5    8
# [3,]    3    6    9


my.matrix <- rbind(c(1:3), c(4:6)) ## binds data row-wise fashion
my.matrix
#      [,1] [,2] [,3]
# [1,]    1    2    3
# [2,]    4    5    6

my.matrix <- cbind(c(1,4), c(2,5), c(3,6)) ## binds data column-wise fashion
my.matrix
#      [,1] [,2] [,3]
# [1,]    1    2    3
# [2,]    4    5    6

Arrays

Multi-dimensional generalization of vectors
Can have more than two dimensions
Created using array() function

my.array <- array(data=1:24,dim=c(3,4,2))
## Each of the two layers constitutes a 3 × 4 matrix as shown in the diagram
# 3 × 4 × 2 array: y * x * z

dim(my.array)
# [1] 3 4 2

my.array[1,2, 1] ## Reads element of first row and second column from 1st layer
my.array[,,2] ## Read all rows and all columns from 2nd layer
my.array[,2,] ## Read second column from both layers
my.array[2, c(1,3), 1] ## Read second row and specific columns (1 and 3) from 1st layer
my.array[2, c(-2,-4), 1] ## Read second row by omitting columns 2 and 4 from 1st layer
my.array[2, 2, 1] <- 32 ## Overwrites the value of second row and second column in 1st layer
my.array[, , -1] ## Read all rows and all columns from by omitting 1st layer

Data Frames

Data frames are used to store tabular data in R
Rectangular data structures with rows and columns
Columns can be of different data types
Created using data.frame() function
It can store numeric data, factor data, and so on

my.dataframe <- data.frame(
    person=c("Peter","Lois","Meg","Chris", "Stewie"),
    age=c(42,40,17,14,1),
    sex=factor(c("M","F","F","M","M"))
    )
## A data frame with the first name, age in years, and sex of five individuals.
#   person age sex
# 1  Peter  42   M
# 2   Lois  40   F
# 3    Meg  17   F
# 4  Chris  14   M
# 5 Stewie   1   M

length(my.dataframe) ## It will return the size of a data frame
# 3

my.dataframe[1,2] ## Reads element of first row and second column
my.dataframe[, c(1,3)] ## Read all rows and specific columns (1 and 3)
my.dataframe[, c(-3)] ## Read all rows by omitting columns 2
my.dataframe$person ## Reading data by using the names of the vectors
my.dataframe[my.dataframe$age>17,] ## Filters data where age is greater than 17 newrecord <- data.frame(person="Brian",age=7, sex=factor("M"))
my.dataframe <- rbind(my.dataframe,newrecord) ## Add a new row by using rbind
funny <- c("High","High","Low","Med","High","Med")
my.dataframe <- cbind(my.dataframe,funny) ## Add a new variable by using cbind

Factors

Used to represent categorical data
Can have levels indicating categories
Useful for ordinal variables
Created using factor() function

funny <- c("High","High","Low","Med","High","Med")
my.factor <- factor(x=funny, levels=c("Low","Med","High"))
# [1] High High Low  Med  High Med
# Levels: Low Med High