titanic.R

# titanic.R

## An R Script on simple exploration of the Titanic dataset
## In RGui, to run an R script's line hold CTRL + R

## Download the dataset into the working directory 

# Check the working directory, 
getwd()

# if you need to change it use 'setwd()'

# Check the files in the directory. 'train.csv' ought to be there
list.files()

# import the data in train.csv into R
df <- read.csv("train.csv")

# Check out the data
dim(df)
str(df)

# Check the variable names
colnames(df)

# Examine the variable 'Embarked'
df$Embarked

# Make a frequency table
table(df$Embarked)

# Create a variable with the 'Age' column
age <- df$Age

# Check the objects in the workspace
ls()

# Compute the arithmetic mean of the ages
mean(age)

# Something is wrong! Check if there are missing values i.e. NAs
anyNA(age)

# Call mean() again, this time removing the NAs
mean(age, na.rm=TRUE)

# Get the range, minimun, maximum, and Tukey's 5-number summary
min(age)
max(age)
range(age)
fivenum(age)
## All functions failed because of the presence of NAs??

# Check the function signature i.e its arguments
args(min)   # This also has the 'na.rm' argument used in 'mean' earlier.

min(age, na.rm=T)

args(range)
range(age, na.rm=T)

table(age)
median(age, na.rm=T)

# Work on the 'Sex' variable
sex <- df$Sex
ls()
table(sex)
anyNA(sex)

# Draw plots
plot(age)  # By default a scatterplot is drawn with numerical data
plot(sex)  # By default, a bar chart is drawn with categorical data
plot(sex, age)  # By default the 2 variables will give a box-and-whiskers plot

# Now, customise the plot, step-by-step for easy understanding
plot(sex, age, col = "red")
plot(sex, age, col = "red", main = "Plot of sex vs. age")
plot(sex, age, col = c("pink", "blue"), main = "Plot of sex vs. age")
plot(sex, age, col = c("pink", "blue"), main = "Plot of sex vs. age", xlab = "Sex", ylab = "Age")

# Draw a histogram
hist(age)

# This is the same chart but now customized.
# Note the styling used to accommodate the many arguments!
hist(age, 
     col = "blue",
     main = "Distribution of Passengers' Age",
     ylab = "No. of Passengers",
     xlab = "Age",
     ylim = c(0, 100)
)