Vectors, Matrices, Lists and Data Frames#

Learning objectives

  • Questions:

    • Vectors in R

    • How to define matrix in R?

    • How to manipulate a data frame in R?

    • How to read text/csv file

  • Objectives:

    • Working with Matrices

    • Creating data frames

    • Importing and exporting data frames

    • Working with text/csv files

  • Keypoints:

    • Working with csv input data

Vector math

  • R can do a lot of mathematical operations on vectors:

a <- 3:7
b <- 20:24
a+b
a>b
a>5
a*b
a/b

  • What if vector lengths don’t match?

a<-1:5
b<-1:6
a
b
a+b

Matrices

  • Vectors are one-dimensional rows of numbers; matrices are two-dimensional tables. Like vectors, all elements of a matrix must be of the same data type.

Let’s create a matrix of zeros:

m <- matrix (0, nrow=3, ncol=4)
m

  • When creating a matrix, you will need to specify number of rows and columns.

  • You can create a matrix from a vector:

m <- matrix(1:12,nrow=3,ncol=4)
m <- matrix(1:12,3,4)
m
dim(m)

  • Another way to create a matrix from a vector:

m <- 1:12
dim(m) <- c(3,4)

Basic matrix math

m1 <- matrix(1:9,nrow=3,ncol=3)
m2 <- matrix(rep(10,9),3,3)
m1
m2
m1+m2
m1*m2
m1 %*% m2

  • Some additional matrix functions:

# Define a matrix
mr <- matrix(runif(9),3,3)
#Transpose a matrix
t(mr)
#Diagonal elements of a matrix
diag(mr)
#Determinant
det(mr)
#Inverse
solve(mr)
solve(mr) %*% mr

Merging Matrices

  • Merging matrices by row and column using rbind and cbind:

m 
m2 <- -1:-12
dim(m2) <- c(3,4)
m2
cbind(m,m2)
rbind(m,m2)

Subsetting

  • Subsetting is extracting elements from vectors/matrices/lists. In R, subsetting is denoted with square brackets: []

str <- c("a", "b","c","d")
str
# Find the subset with index 1 for str:
str[1]
# Find the subset with index 2:4 for str:
str[2:4]

  • You can see that we can use a vector to subset another vector.

  • Subsetting is very flexible. You can use a vector of logical values to subset another vector:

a <- c (1, 2, 3, 4, 3, 2, 1)
a
a>2
a[a>2]

  • You can use this method to filer out elements which are NaNs:

a <- c(1:5,NaN,TRUE)
a
# Find the location of *NaN* value using `is.nan()` function
ind <- is.nan(a)
ind
# Subset with location of *NaN* value
a[ind]
a[is.nan(a)]
# Subset with location of `Not NaN` values using `!`
a[!ind]

  • Subsetting matrices

    • When subsetting a matrix, we use two indices (row and column):

m <- matrix(1:12,nrow=3,ncol=4)
m
m[2,3]
m[1,1:3]

  • Extracting a whole row or column:

m[2,]
m[,4]

Lists

  • In matrices and vectors, all elements belong to the same class. A collection of variables from different classes is called a list.

str <- "Clemson"
a <- 5
b <- 4L
list1 <- list(str,a,b)
list1

  • Lists are very flexible. A vector, or a matrix, can be an element of a list:

c <- 6i ^ (-3:3)
d <- 1:10 < 5
list2 <- list(str,a,b,c,d)
list2

  • Subsetting lists

    • Elements of a list can also be extracted by subsetting, using the dollar sign $:

str <- c("a", "b","c","d")
list1 <- list(l1=str,l2=4:6)
list1
# Use $ to call a variable name
list1$l1[3]

Data Frames

  • Data frames are used to store tables, where columns correspond to a particular variable, and rows correspond to a particular observation.

  • Data Frame characteristics:

    • Column name should not be empty

    • Row name should be unique

    • Data can be numeric, integer, character

    • Each column contains same number of data items

df <- data.frame(data=sample(12),title=LETTERS[sample(12)])
dim(df)
nrow(df)
ncol(df)

  • There are many readily available data frames in R, for example iris data set:

data(iris)
dim(iris)
head(iris)

Names of objects in data frames

  • Using name() function:

names(iris)
# Change name
names(iris) <- c("a", "b", "c","d","e")
head(iris)

#Change name for particular columns:
names(iris)[4] <- "new_name"

Getting data from data frames

  • Columns of a data frame could be extracted with $ (as in a list) or with [] (as in a matrix).

data(mtcars)
head (mtcars)
mpg1 <- mtcars$mpg
mpg2 <- mtcars[,1]
cylinder <- mtcars$cyl

Reading and writing tables

  • Reading table

    • Data frames can be created by reading from a file (a text file, or a CSV / Excel file). The file can be on a local computer, or online.

    • read.table: read table in text format

    • read.csv: read table in csv format

    • read.xlsx: read table in excel format (require xlsx packages)

    • readLines: read lines of a text file

  • Writing table

    • In a similar fashion, a data frame can be saved as a text/CSV/Excel file:

    • write.table: write table in text format

    • write.csv: write table in csv format

    • write.xlsx: write table in excel format (require xlsx packages)

    • writeLines: write lines of a text file

  • In this example, I perform reading online sale data and save the output to current working directory:

# Read online csv data
saledata <- read.csv('https://support.spatialkey.com/wp-content/uploads/2021/02/Sacramentorealestatetransactions.csv')
dim(saledata)
names(saledata)

# Save output to csv file
write.csv(saledata,'SaleData.csv')

  • Here, we read a poem from an online text file, and save ten lines in the working directory:

poem <- readLines("http://lib.ru/SHAKESPEARE/sonnets.txt")
poem[10:20]
writeLines(poem[10:20],"Sonnet1.txt")