library(tidyverse)
library(titanic)
options(digits = 3)   # report 3 significant digits

titanic <- titanic_train %>%
  select(Survived, Pclass, Sex, Age, SibSp, Parch, Fare) %>%
  mutate(Survived = factor(Survived),
         Pclass = factor(Pclass),
         Sex = factor(Sex))

── Attaching packages ─────────────────────────────────────── tidyverse 1.3.1 ──

✔ ggplot2 3.3.5     ✔ purrr   0.3.4
✔ tibble  3.1.3     ✔ dplyr   1.0.7
✔ tidyr   1.1.3     ✔ stringr 1.4.0
✔ readr   2.0.0     ✔ forcats 0.5.1

── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
✖ dplyr::filter() masks stats::filter()
✖ dplyr::lag()    masks stats::lag()

This first analysis uses only the following informations from the train data:
    - Survived: Passenger Survival Indicator
    - Pclass: Passenger Class
    - Sex: Sex
    - Age: Age
    - SibSp: Number of Sibling/Spouses Aboard
    - Parch: Number of Parents/Children Aboard
    - Fare: Passenger Fare

titanic %>% filter(!is.na(Age)) %>%
    ggplot(aes(Age, fill = Sex)) +
    geom_density(alpha = 0.3)

The graph of density of age grouped by sex indicates that both sex had same general shape of age distribution.
It also reveals the distribution is bimodal, with one mode around 25 years of age and a second smaller mode around 5 years of age.

params <- titanic %>%
  filter(!is.na(Age)) %>%
  summarize(mean = mean(Age), sd = sd(Age))

titanic %>% filter(!is.na(Age)) %>%
  ggplot(aes(sample = Age)) +
  geom_qq(dparams = params) +
  geom_abline()

The quantile-quantile plot shows how well the passengers age approximates to a normal distribution, therefore it can be described by its mean value and standard deviation.

titanic %>% filter(!is.na(Survived) & !is.na(Sex)) %>%
  ggplot(aes(x = Survived, fill = Sex)) +
  geom_bar(position = position_dodge())

This barplot shows the survivors count (1 being those who survived) grouped by sex. One can conclude that:
    - less than half of the passengers survived;
    - most of the male did not survive;
    - most of the female survived;
    - most of the survivors were female.

titanic %>% filter(!is.na(Age)) %>%
  ggplot(aes(Age, y = ..count.., fill = Survived)) +
  geom_density(alpha = 0.2)

Here is presented a density plot of age filled by survival status. Through its analysis it is possible to conclude that:
    - In the age group of 0-8 it was more likely to survive than die;
    - The age group of 18-30 had the most deaths;
    - The age group of 70-80 had the highest proportion of deaths.

titanic %>% filter(!Fare == 0) %>%
  ggplot(aes(Survived, Fare)) +
  geom_boxplot() +
  scale_y_continuous(trans = "log2") +
  geom_jitter(width = 0.1, alpha = 0.2)

The bloxplot illustrates the passengers fare grouped by survival status, together with all the data points. The fares are presented in log2 scale.
It is possible to conclude that:
    - Passengers who survived generally payed higher fares than those who did not survive;
    - The interquartile range for fares was bigger for passengers who survived;
    - The median fare was lower for passengers who did not survive;
    - Three individuals paid a fare around $ 500 and all three survived;
    - Most individuals who paid a fare around $ 8 did not survive.

titanic %>% ggplot(aes(x = Pclass, fill = Survived)) +
  geom_bar(position = position_dodge())

titanic %>% ggplot(aes(x = Pclass, fill = Survived)) +
  geom_bar(position = position_fill())

titanic %>% ggplot(aes(x = Survived, fill = Pclass)) +
  geom_bar(position = position_fill())

The first two barplots above show the passenger class filled by survival status, one in total counts and one in relative proportions. The third barplot switches the variables and shows the survival status filled by passenger class, in proportion.
One can conclude that:
    - There were more third class passengers than passengers in the first and second classes combined;
    - There were fewer passengers from the second class;
    - Survival proportion was highest for first class passengers, followed by second class. Third-class had the lowest survival proportion;
    - Most passengers in first class survived. Most passengers in other classes did not survive;
    - The majority of those who did not survive were from third class.

titanic %>% filter(!is.na(Age)) %>%
  ggplot(aes(Age, y = ..count.., fill = Survived)) +
  geom_density(alpha = 0.2) +
  facet_grid(Pclass ~ Sex)

Here is presented a grid of density plots for age, filled by survival status, faceted by sex and passenger class.
The analysis indicates that:
    - The largest group of passengers was third-class males;
    - The age distribution differs across passenger classes;
    - The gender distribution differs across passenger classes;
    - Most first-class and second-class females survived;
    - Almost all second-class males did not survive, with the exception of children.