Data Structures

Data Structures#

Learning objectives

  • Questions:

    • Data types / classes in R

    • Random numbers

    • Vectors

    • Missing values

  • Objectives:

    • 5 data types / classes of objects

    • Working with vectors

  • Keypoints:

    • Classes of objects

    • Vectors

Classes of objects

  • In R, there are 5 main classes of objects:

    • characters a, b

    • numeric: 2.3

    • integer: 5 or 5L

    • complex: 2+3i #consists of real and imaginary parts

    • logical: TRUE/FALSE

str <- "string"
class(str)
a <- 5
class(a)
b <- 4L
class(b)
c <- 1+6i 
class(c)
d <- b < 5
class(d)

Logical operations

  • Logical operations and and or are computed with && and ||:

a<-TRUE
b<-FALSE
a&&b
a||b

  • Logical not is computed with !:

!a
!b
a && !b

Numbers

  • In R, a number is considered a numeric type:

e <- 5
class(e)

  • To get an integer, insert L as suffix

f <- 5L
class(f)

  • Special number: Inf: infinity

g<-5/0
class(g)
5/0
-5/0
log(0)

  • NaN (Not a Number) specifies an undefined value:

h <- 0/0
h
log(-1)
asin(4)

  • Sometimes, when you are dealing with data, you have missing values. In R, there is a special way to denote missing values: NA (“not available”). NaN is a special type of NA.

Random Number & seed

  • Create random numeric numbers using runif

runif(1)
runif(3)
runif(2,10,20)

  • Set the seed of R’s random number generator, which is useful for creating simulations or random objects that can be reproduced.

set.seed(1234)
runif(3)

  • Create a sample of random numbers:

sample(12,5)
sample(12)
sample(letters,4)

  • sample returns a set of variables, rather than one variable. Sets of variables of the same type are called vectors.

Vectors

  • Vectors are sets of variables of the same class, in a certain order. The simplest, and a very useful, type of a vector is a range of values, specified with a colon:

x <- 1:5
x <- -3:3

  • The same could be accomplished with the seq function:

seq (from=1, to=5)
seq (from=1, to=5, by=2)
seq (1, 5, 2)

  • You can also define a vector by combining a bunch of values (c stands for “combine”):

str <- c("a", "b", "c")
a   <- c(4, 5.6, 20)
b   <- c("TRUE", "FALSE")

  • To create a vector where all elements have the same value, use the rep function (for “repetition”):

x <- rep (3, 5)

  • If a vector is combined from variables of different types, R tries to coerce them into the same type. For example, logical values can be coerced into numbers (TRUE becomes 1, and FALSE becomes 0):

c(TRUE, FALSE)
c(TRUE, FALSE, 5)

  • Likewise, numbers can be coerced into characters:

c (5, "a")

  • In this example, everything is ultimately coerced into characters:

str1 <- c("a", "b", TRUE, 5, 4.5)
str1
class(str1)

Explicit Coercion

  • Convert objects from one class to another, using as. function:

a <- 0:5
class(a)
as.numeric(a)
as.logical(a)
as.character(a)

  • How about Nonsensical Coercion?

str <- c("a","b","c")
class(str)
as.numeric(str)
as.logical(str)
as.character(str)

Missing values

  • In order to detect missing values or bad values (NaN, NA, Inf), we can use these functions:

    • is.na() test NA value

    • is.nan() test NaN value

    • is.infinite() test Inf value

    • NaN is NA but the reverse is false

v <- c(TRUE, 6, 1/0,NA, NaN,-6/0)
v
is.na(v)
is.nan(v)
is.infinite(v)