gistfile1.txt

# DATA FRAME OPERATIONS IN R

# Create data frame
# A dataset is ~ table (list of vectors)
id <- c(1,2,3)
name <- c("John", "Kirk", "AJ")
age <- c(21,27,18)
employees <- data.frame(ID=id, Name=name, Age=age)
employees

city <- c("New York","Chicago","London")
address <- data.frame(ID=id, City=city)
address

more.id <- c(11,12,13)
more.name <- c("Kira", "Jen", "Liz")
more.age <- c(25,27,21)
more.employees <- data.frame(ID=more.id, Name=more.name, Age=more.age)
more.employees

# ----------------------------------
# Inspect data frames

# check first few rows
head(employees)

# check some last rows
tail(employees)

# ----------------------------------

# Accessing elements of data frame
# data frames are addressed by row and columns in the matrix notation

# get a value from a cell (a particular row and a particular column)
employees[1,2] # first row, second column
employees[1,"Name"] # first row, column by name
employees[1,]$Name # first row, column by name

# get one row
employees[1,]

# get one column
employees[,2]
employees[,"Name"]
employees$Name

# get multiple rows/columns (subset)
employees[1:2,] # returns 2 rows
employees[,1:2] # returns 2 columns
employees[,c(1, 2)] # returns 2 columns
employees[,c("ID", "Name")] # returns 2 columns

# get rows that pass a test
employees[employees$Age > 20, ]
# ----------------------------------

# Data Frame properties

# number of rows
nrow(employees)

# number of columns
ncol(employees)

# summary stats
summary(employees)

# structure
str(employees)

# ----------------------------------

# Manipulate data frame

# set value
employees[3,"Age"] <- 29

# order
employees[order(employees$Age),]

# reverse order
employees[order(employees$Age, decreasing=T),]

# merging data frames
merge(employees, address, by="ID")

# add rows
all.employees <- rbind(employees, more.employees)
all.employees

# add columns
cbind(employees, city) # city is treated as a data frame

# grouping
# aggregate is similar to group by in SQL. Here are the # employees grouped by age
aggregate(all.employees[,2], list(Age=all.employees$Age), FUN=length)

# A column and a row of a data frame is a vector and all vector operations can be applied to it e.g. math/stats functions 
mean(all.employees$Age)
# ----------------------------------

# Test for data frame
is.data.frame(employees)