Indexing
04 Feb 2022On Thursday Problem Set 2 was distributed. A digital version of the prompt can be accessed here: ProblemSet2.pdf. The highlights from this week are summarized below:
Indexing
To extract individual elements or sub-vectors from a vector the [] notation can be used. Numbers (integers) supplied inside of the [] will instruct R to extract the objects located at the corresponding position inside of the vector.
# extracts the element at position 26 of the vector letters
# will return "z"
letters[26]
# will return "h"
letters[8]
# will return vector containing "h" and "z"
letters[c(8, 26)]
# will return vector containing "a", "b", and "c"
letters[1:3]
# will return letter "i"
letters[8+1]
# will return "d"
Fourth_letter <- 4
letters[Fourth_letter]
# will return vector of length 24 containing
# all letters except "a" and "h"
letters[-c(1,8)]
# assignes the first three letters to a
# new vector called x
x <- letters[1:3]
# replaces the second element of vector x
# with the word "Hello"
x[2] <- "Hello"
We also discussed the which() function which returns a vector of index locations meeting a logical condition and the subset() function which can be used to extract elements meeting a certain condition.
# evaluates to 8 and 17
which(letters == "h" | letters == "q")
# extract letters h and q
subset(x = letters, letters == "h" | letters == "q")
Indexing w/ TRUE and FALSE
Among the things covered was the use of TRUE and FALSE inside of the [] brackets to extract elements from vectors.
abc <- c("a", "b", "c", "d", "e")
# returns a vector containing all the elements of abc
abc[c(TRUE, TRUE, TRUE, TRUE, TRUE)]
# returns a vector that is empty (character(0))
abc[c(FALSE, FALSE, FALSE, FALSE, FALSE)]
# the empty vector is of type character
typeof(x = abc[c(FALSE, FALSE, FALSE, FALSE, FALSE)])
# the empty vector is indeed empty and has length 0
length(x = abc[c(FALSE, FALSE, FALSE, FALSE, FALSE)])
# returns a vector containing only the second and forth element of abc
# that is "b" and "d"
abc[c(FALSE, TRUE, FALSE, TRUE, FALSE)]
# creates an empty vector of type double
foo <- c(pi, -0.5)[c(FALSE, FALSE)]
# as does this
foobar <- pi[FALSE]
With this in mind, we can use logical arguments to subset vectors.
# returns a vector of length 26
# containing the logical value FALSE 24 times and
# the logical value TRUE 2 times (in the first and
# last position)
letters == "a" | letters == "z"
# returns the letters "a" and "z"
letters[letters == "a" | letters == "z"]
Managing Objects and Data
R is perfectly capable of serving as a file-system browser.
# the present working directory can be identified via
getwd()
# files and folders present in the current working
# directory can be identified via
list.files()
# files and folders present in an arbitrary
# directory can be identified by specifying the path
# for example:
list.files(path = "C:/Users/UserName/Desktop") # windows
list.files(path = "/Users/Username/Desktop") # mac
# the working directory can be changed via
setwd(dir = "C:/Users/UserName/Desktop") # windows
setwd(dir = "/Users/Username/Desktop") # mac
# saving objects stored in R's active memory
# in the current working directory (with the .Rdata extension)
x <- rnorm(n = 100, mean = 0, sd = 10)
y <- runif(n = 55, min = 0, max = 1)
save(x, y, file = "my-data.RData")
# removes all objects in R's active memory
rm(list = ls())
# load the objects x and y back into R's active memory
load(file = "my-data.RData")