R_Basics.R

# ==========
# = Basics =
# ==========

# Mathematical operations
# Parentheses: ()
# Exponent:    ^ or **
# Multiply:    *
# Divide:      /
# Addition:    +
# Subtraction: -
# R follows PEMDAS order of oprations

# Assignment operator: <-
# variable <- value to be assigned (order matters!)

x <- 1/40  # Assign x to the value of 0.025
x <- x + 1 # Increment x

# ===========
# = Vectors =
# ===========

1:5 # will print 1 2 3 4 5

# Can assign vector to variable
var <- 1:5

# Can use vectors to loop or create functions
2^(1:5) # will print 2 4 8 16 32
2^var   # will also print 2 4 8 16 32

# Can use 2 vectors at the same time
y <- 1:4
z <- 5:8
y + z # will print 6 8 10 12

# ======================
# = Variable Managment =
# ======================

# ls() lists variables being used
ls()

# rm() removes a variable from the environment
# rm("variable_name") works the same as rm(variable_name)
x2 <- x
rm(x)
rm("x2")

# One option for rm is list
# List will remove all variables in that list
# rm(list = ls()) will remove all current variables
# rm(list = c("var1", "var2", "var3")) will remove var1, var2, var3

# ======================
# = Package Management =
# ======================

# To find out what packages you have installed, use
# installed.packages()
installed.packages()

# To install a package, use
# install.packages("package_name")
# Each package will only need to be installed on your computer once

install.packages("ggplot2")
install.packages("dplyr")
install.packages("gapminder")

# To update a package, use
# update.packages("package_name")

# To load a package, use
# library("package_name")
# Packages will need to be loaded in each script

library("ggplot2")
library("dplyr")
library("gapminder")

# ================
# = Getting Help =
# ================

# To get more info about a function, use
# ?function_name
# or
# help(function_name)

?log
help(exp)

# If you don't know the exact name of the function, use
# ??function_name_ish

??read.tab # returns information for functions
           #    readr::read_table
           #    base::base-defunct
           #    utils::read.table <= yay!

# You can use ? operator to get information about 
# operators as well

?"<-" # returns info about the assignment operator

# To get information about a package, use
# vignette("package_name")
# using vignette() will show lottttttts of info

vignette("dplyr") # shows info about the dplyr package

# ===================
# = Data Structures =
# ===================

# Cats Data Frame Example
# Note: c() is for combine

cats <- data.frame(coat = c("calico", "black", "tabby"),
                   weight = c(2.1, 5.0, 3.4),
                   likes_string = c(1, 0, 1))
cats # will show data frame table

# View prettier table by double-clicking "cats" in 
# environment panel or use
# View(data_frame_name)

View(cats)

# ===============================
# = Reading and Writing to File =
# ===============================

# To output cat data to a .csv file, use
# write.csv() 

write.csv(cats, "data/meow.csv")

# To read this back, use 
# read.csv()

felines <- read.csv("data/meow.csv")

# Note that read.csv() creates a new variable based on
# row numbers in the data.

# Based on the above, our cats data and our felines data
# has the same information, but cats has 3 variables
# and felines has 4 variables (3 + row numbers)

# To print a row of data, use 
# data_frame_name$row_name

cats$weight  # displays weight row of data as a vector
felines$coat # displays coat row of data as a vector

# Can perform operations on data in the data frame
# Example: Convert weight from kg to lbs
cats$weight <- cats$weight * 2.2

# ==============
# = Data Types =
# ==============

# To see the data type of a variable, use
# class(variable_name)

class(cats$coat)   # will display "factor"
class(cats$weight) # will display "numeric"
class(cats)        # will display "data.frame"

# To see the data, data type within a data frame, use
# structure function:
# str(variable_name)

str(cats)

# ================
# = More Vectors =
# ================

num_vector <- c(1, 3, 5)
chr_vector <- c("a", "b", "c")
chr_vector2 <- c("d", "e", "f")

# combining vectors of characters creates a longer 
# character vector

comb_vector <- c(chr_vector, chr_vector2)

# combining vectors of numbers adds the numbers together

num_vector2 <- c(100, 10, 20)
num_vector + num_vector2 # will display (101, 13, 25)

# creating numerical vectors
# can use : and seq()

my_series <- 1:10 
my_series2 <- seq(10)

# seq() can have a different step, or spacing, between
# elements
new_step <- seq(from = 1, to = 10, by = 0.1)

# can name each item in a vector
named_vector <- 5:8                    # original vector
naming_vector <- c("a", "b", "c", "d") # vector of names
names(named_vector) <- naming_vector   # combining them
named_vector                           # display!

# When adding number vectors of different lengths,
# the shorter vector repeats itself until it is as long
# as the longer vector, and then they add.
long_num_vector <- 0:9
short_num_vector <- 0:1
long_num_vector + short_num_vector

# something different happens when naming
# after the end of the naming vector, names are blank
larger_than_alphabet <- 1:30
names(larger_than_alphabet) <- LETTERS
larger_than_alphabet

# =============
# = Real Data =
# =============

# load it from csv
gapminder <- read.csv("data/gapminder_data.csv")

# alternative way to load the data
library(gapminder) # load the package
data("gapminder")  # load data frame called "gapminder"
attach(gapminder)  # copy gapminder into our environment

# learn about data
str(gapminder)      # types of data in the data frame
nrow(gapminder)     # number of rows of data
ncol(gapminder)     # number of columns of data
dim(gapminder)      # alternative way to see dimensions 
                    # of frame: nrow <tab> ncol
colnames(gapminder) # names of columns; variable names

# ================
# = Data Subsets =
# ================

x <- c(5.4, 6.2, 7.1, 4.8, 7.5)
names(x) <- letters[1:5] # Using [] selects part of a variable or data frame
x

x[1]        # returns first value of x (a 5.4)
x[3]        # returns third value of x (c 7.1)
x[c(1, 3)]  # returns first and third value of x 
            #   (a 5.4 c 7.1)
x[1:4]      # returns all values of x from 1 to 4
            #   (a 5.4 b 6.2 c 7.1 d 4.8)
x[6]        # returns nothing; no sixth value
x[-2]       # returns all of x except second element
x[-(2:4)]   # returns all of x except element 2-4

# head shows the first 6 rows of the table and variables
# can change to a different number of rows if necessary
head(gapminder)         # shows 
head(gapminder["pop"])  # shows first 6 populations
head(gapminder[,5])     # shows first 6 rows of 5th col

# First element of square braces determine rows [rows, columns]
gapminder[3,] # show all columns of the third row

# Show the life expectancy on line 138
gapminder[138,"lifeExp"]

# Show first six rows of the double type columns
head(gapminder[c(4, 6)])
gapminder[1:6, c(4, 6)]
gapminder[1:6, c("lifeExp", "gdpPercap")]

# Subset into a country
albania <- gapminder[13:24, 1:6]

# Exercise: 
#   Find a subset for Afghanistan
#   Add a new column to that subset for GDP
#   Calculate GDP by multiplying population by GDP Per Capita
#   Save new subset into a .csv file in your data folder
afghanistan <- gapminder[1:12, ]
afghanistan$gdp <- afghanistan$gdpPercap * afghanistan$pop
View(afghanistan) # Check out what we've done to confirm
write.csv(afghanistan,"data/results.csv")

# =========
# = Plots =
# =========

# Plot Prep - making sure your data is present in your environment
library(ggplot2)
library(dplyr)
install.packages("tidyr")
library(tidyr)
install.packages("knitr")
library(knitr)
gapminder <- read.csv("data/gapminder_data.csv")
head(gapminder)

# Troubleshooting 
getwd()          # find current working directory
setwd("./data/") # change working directory to where your data is
                 # Note that "./data/" is a relative path, so your setwd() path
                 # may be different based on your current working directory

# Plot command
# Set data to your data frame
# Second factor - aes is for aesthetics
# geom_point() adds your data to the plot in the form of scatterplot
# geom_line() adds data to the plot in the form of a line graph

ggplot(data = gapminder, 
       aes(x = gdpPercap,     # set x axis
           y = lifeExp,       # set y axis
           color = continent, # set color by continent data
           by = country       # set line by country
           )) + geom_line()   # display data in a line graph
          + geom_point()      # add another layer - scatterplot

# Layering plots
# Layers one on top of another, same aesthetics

ggplot(data = gapminder, aes(x = gdpPercap, y = lifeExp, color = continent, 
           by = country)) + geom_line() + geom_point() 

# aes, color specific to different plots is possible

ggplot(data = gapminder, aes(x = gdpPercap, y = lifeExp, by = country)) + 
  geom_line(aes(color = continent)) + geom_point(color = "blue")

# Change to logarithmic scale by adding a scale_x_log10() function
#   Alpha changes the points' transparency, which helps when points land on top
#   of each other
#   Alpha scale: 1.0 is solid; 0.0 is completely transparent / invisible

ggplot(data = gapminder, 
       aes(x = gdpPercap, y = lifeExp, color = continent, by = country)) + 
  geom_point(alpha = 0.5) + 
  scale_x_log10()

# Add a regression line using geom_smooth()
# lm = linear model
# gray lines around the regression shows a confidence interval

ggplot(data = gapminder, 
       aes(x = gdpPercap, y = lifeExp, color = continent)) + 
  geom_point(alpha = 0.5, aes(shape = continent)) + 
  scale_x_log10() +
  geom_smooth(method = "lm")

# Remove a legend by doing show.legend = FALSE at the layer you're 
# most interested in

ggplot(data = gapminder, 
       aes(x = gdpPercap, y = lifeExp, color = continent)) + 
  geom_point(alpha = 0.5, aes(shape = continent), show.legend = FALSE) + 
  scale_x_log10() +
  geom_smooth(method = "lm")

# turn a legend off completely in theme


# clean up for publication
# add scale_y_continuous() with options for y scale that allow you to change
#   to percent and change the intervals
# add theme_bw() to remove background color
# add ggtitle() to add a title
# add xlab() to add a label to the x axis
# add ylab() to add a label to the y axis

ggplot(data = gapminder, 
       aes(x = gdpPercap, y = lifeExp, color = continent)) + 
  geom_point(aes(shape = continent), size = 2, alpha = 0.25) + 
  scale_x_log10() +
  geom_smooth(method = "lm") +
  scale_y_continuous(limits = c(0,100), breaks = seq(0, 100, by = 10)) +
  theme_bw() +
  ggtitle("Interaction of GDP Per Capita and Life Expectancy") +
  xlab("GDP Per Capita ($)") +
  ylab("Life Expectancy (Years)")

# Export the plot
ggsave(file = "life_expectancy.png")
ggsave(file = "life_expectancy.pdf")

# Troubleshooting the images

ggsave(file = "life_exp.png", width = 20, height = 15, units = "cm")

# Boxplot example

ggplot(data=gapminder, aes(x = continent, y = lifeExp)) + geom_boxplot() +
  geom_jitter(alpha = 0.5, color = "tomato")

# ================
# = Conditionals =
# ================

# If / Else Statements

number <- 37
if (number > 100) {
  print("ERMAGERD bigger than 100!")
} else {
  print("Not so big")
}

# Comparison
# Greater than:             >
# Greater than or equal to: >=
# Equal to:                 ==
# Less than:                <
# Less than or equal to:    <=
# Not equal to:             !=

# If / Else If / Else Chains
# Else is a good catch-all or default statment

number <- 37
if (number > 0) {
  print(1)
} else if (number < 0) {
  print(-1)
} else {
  print(0)
}

# Loops
# Each of the first two loops print the numbers 1-10

numbers <- 1:10
for (number in numbers) {
  print(number)
}

for (i in 1:10) {
  print(i)
}

for (i in letters) {
  print(i)
}

# sum
sum <- 0
vector <- c(4, 8, 15, 16, 23, 42)
for (i in vector) {
  sum -> sum + i
}
print(sum)

# =====================
# = Writing Functions =
# =====================

# Reading R's Functions
# Call a function without arguments and it will display 
# the function's source code

nrow # displays source code for nrow

# Function format
# name_of_function <- function(variables_function_needs) { 
#   function_here 
#   return(data_to_return)
# }

# Example
fahr_to_kelvin <- function(f_temp){
  kelvin <- ((f_temp - 32) * (5 / 9)) + 273.15
  return(kelvin)
}

# Make sure to call your function after you define it
# Otherwise, R won't know what your function does
# If it's not in your Environment list with your variables,
# R doesn't know it exists

# Boiling point
fahr_to_kelvin(212) # returns 373.15, the boiling point in kelvin

cel2fahr <- function(cel) {
  fahr <- (cel * 9 / 5) + 32
  return(fahr)
}

cel2fahr(70)

# =================
# = Markdown in R =
# =================

install.packages(c("rmarkdown", "formatR"))
library(dplyr)
library(gapminder)
library(tidyr)
library(knitr)
library(rmarkdown)
library(formatR)