In [11]: # Titanic EDA Assignment - Final Version

import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
from scipy import stats
from scipy.stats import chi2_contingency

# Load the dataset

df = pd.read_csv('titanic_train_.csv') # Update path if needed

# 1. Summary Table
summary = []
for col in df.columns:
dtype = df[col].dtype
distinct = df[col].nunique(dropna=False)
if pd.api.types.is_numeric_dtype(df[col]):
mean = df[col].mean()
median = df[col].median()
std = df[col].std()
rng = (df[col].min(), df[col].max())
else:
mean = median = std = None
rng = (None, None)
summary.append({
'Attribute': col,
'Type': str(dtype),
'Distinct': distinct,
'Mean': mean,
'Median': median,
'Std': std,
'Range': f"{rng[0]} – {rng[1]}"
})

summary_df = pd.DataFrame(summary)
display(summary_df)
 Attribute Type Distinct Mean Median Std Range

0 PassengerId int64 891 446.000000 446.0000 257.353842 1 – 891

1 Survived int64 2 0.383838 0.0000 0.486592 0–1

2 Pclass int64 3 2.308642 3.0000 0.836071 1–3

None –
3 Name object 891 NaN NaN NaN
None

None –
4 Sex object 2 NaN NaN NaN
None

0.42 –
5 Age float64 89 29.699118 28.0000 14.526497
80.0

6 SibSp int64 7 0.523008 0.0000 1.102743 0–8

7 Parch int64 7 0.381594 0.0000 0.806057 0–6

None –
8 Ticket object 681 NaN NaN NaN
None

0.0 –
9 Fare float64 248 32.204208 14.4542 49.693429
512.3292

None –
10 Cabin object 148 NaN NaN NaN
None

None –
11 Embarked object 4 NaN NaN NaN
None

In [14]: import pandas as pd

import matplotlib.pyplot as plt
import seaborn as sns
from scipy import stats

# Set plot style

sns.set(style="whitegrid")

# Histogram of Age
plt.figure(figsize=(6, 4))
df['Age'].hist(bins=20, color='skyblue', edgecolor='black')
plt.title('Histogram of Age')
plt.xlabel('Age')
plt.ylabel('Count')
plt.grid(True)
plt.tight_layout()
plt.show()

# Boxplot of Fare
plt.figure(figsize=(6, 4))
df.boxplot(column='Fare', grid=False)
plt.title('Boxplot of Fare')
plt.ylabel('Fare')
plt.tight_layout()
plt.show()

# Violin Plot of Age by Survival

plt.figure(figsize=(6, 4))
sns.violinplot(data=df, x='Survived', y='Age', palette='Set3')
plt.title('Age Distribution by Survival')
plt.xlabel('Survived (0 = No, 1 = Yes)')
plt.ylabel('Age')
plt.tight_layout()
plt.show()

# Scatter Plot: Age vs. Fare

plt.figure(figsize=(6, 4))
plt.scatter(df['Age'], df['Fare'], alpha=0.5, c='green')
plt.title('Age vs. Fare')
plt.xlabel('Age')
plt.ylabel('Fare')
plt.grid(True)
plt.tight_layout()
plt.show()

# QQ Plot of Age
plt.figure(figsize=(6, 4))
stats.probplot(df['Age'].dropna(), dist="norm", plot=plt)
plt.title('QQ Plot of Age')
plt.grid(True)
plt.tight_layout()
plt.show()

# Correlation Heatmap
plt.figure(figsize=(6, 5))
corr = df[['Survived', 'Age', 'Fare', 'Pclass', 'SibSp', 'Parch']].corr()
sns.heatmap(corr, annot=True, cmap='coolwarm', fmt=".2f")
plt.title('Correlation Heatmap')
plt.tight_layout()
plt.show()

# Pair Plot of Numeric Features Colored by Survival

clean_df = df[['Survived', 'Age', 'Fare', 'Pclass']].dropna()
sns.pairplot(clean_df, hue='Survived', palette='husl')
plt.suptitle('Pair Plot of Numeric Features Colored by Survival', y=1.02)
plt.tight_layout()
plt.show()
<ipython-input-14-c9bafd225947>:31: FutureWarning:

Passing `palette` without assigning `hue` is deprecated and will be removed

in v0.14.0. Assign the `x` variable to `hue` and set `legend=False` for the
same effect.

sns.violinplot(data=df, x='Survived', y='Age', palette='Set3')

Histogram (Age):
The Age distribution is right-skewed, with most passengers between 20–40
years. Some very young (babies) and very old passengers are present.

Boxplot (Fare):
Shows many outliers, especially in the higher fare range. Median Fare is low,
indicating most passengers paid a small amount.

Violin Plot (Age vs. Survived):

Survivors include a higher proportion of young children and some older adults.
Non-survivors were mostly aged 20–40.

Scatter Plot (Age vs. Fare):

 No clear linear relationship, but many survivors cluster in low age and moderate
fare zones.

QQ Plot (Age):
Age does not follow a perfect normal distribution; tails deviate from the normal
line.

Correlation Heatmap:
Positive correlation between Fare and Survived. Negative correlation between
Pclass and Survived (i.e., lower class = lower survival).

Pair Plot:
Survivors tend to be in higher classes (Pclass=1) and paid higher fares.

In [18]: import pandas as pd

from scipy.stats import chi2_contingency

# Categorical features to analyze

categorical_features = ['Sex', 'Pclass', 'Embarked', 'SibSp', 'Parch']

# Store chi-square test results

chi_square_summary = []

print(" 🔍
Chi-Square Test Results:\n")
for feature in categorical_features:
# Create contingency table
contingency_table = pd.crosstab(df[feature], df['Survived'])

# Perform Chi-square test

chi2_stat, p_val, dof, expected = chi2_contingency(contingency_table)

# Append results
chi_square_summary.append({
'Feature': feature,
'Chi-Square': round(chi2_stat, 2),
'p-value': round(p_val, 4),
'DoF': dof,
'Significant (< 0.05)': 'Yes' if p_val < 0.05 else 'No'
})

# Convert results to DataFrame

chi_square_df = pd.DataFrame(chi_square_summary)
display(chi_square_df)

# -------------------------------
# Survival Rate by Each Feature

📊
# -------------------------------
print("\n Survival Rates by Category:\n")
 for feature in ['Sex', 'Pclass', 'Embarked']:
survival_rate = df.groupby(feature)['Survived'].mean().reset_index()
survival_rate.columns = [feature, 'Survival Rate']
print(f"\nSurvival Rate by {feature}:")
display(survival_rate)

🔍 Chi-Square Test Results:

Feature Chi-Square p-value DoF Significant (< 0.05)

0 Sex 260.72 0.0000 1 Yes

1 Pclass 102.89 0.0000 2 Yes

2 Embarked 26.49 0.0000 2 Yes

3 SibSp 37.27 0.0000 6 Yes

4 Parch 27.93 0.0001 6 Yes

📊 Survival Rates by Category:

Survival Rate by Sex:
Sex Survival Rate

0 female 0.742038

1 male 0.188908

Survival Rate by Pclass:

Pclass Survival Rate

0 1 0.629630

1 2 0.472826

2 3 0.242363

Survival Rate by Embarked:

Embarked Survival Rate

0 C 0.553571

1 Q 0.389610

2 S 0.336957

Tested the relationship between categorical variables and Survived.

Significant associations (p < 0.05) found with:

Sex: Strong relation with survival; females survived more.

Pclass: Higher classes had higher survival.
Embarked: Slight effect; passengers from Cherbourg (C) had higher survival.
 SibSp: Moderate impact; people traveling with 1–2 family members had
better chances.

No significant association found with:

Parch: Parental status didn’t independently affect survival.

In [19]: # Set plot style

sns.set(style="whitegrid")

# 1. Survival Count Plot

plt.figure(figsize=(6, 4))
sns.countplot(x='Survived', data=df, palette='Set2')
plt.title("Survival Distribution")
plt.xticks([0, 1], ['Did Not Survive', 'Survived'])
plt.show()

# 2. Survival by Sex
plt.figure(figsize=(6, 4))
sns.barplot(x='Sex', y='Survived', data=df, palette='pastel')
plt.title("Survival Rate by Sex")
plt.ylabel("Survival Rate")
plt.show()

# 3. Survival by Passenger Class

plt.figure(figsize=(6, 4))
sns.barplot(x='Pclass', y='Survived', data=df, palette='muted')
plt.title("Survival Rate by Passenger Class")
plt.ylabel("Survival Rate")
plt.show()

# 4. Survival by Embarked Port

plt.figure(figsize=(6, 4))
sns.barplot(x='Embarked', y='Survived', data=df, palette='coolwarm')
plt.title("Survival Rate by Embarked Port")
plt.ylabel("Survival Rate")
plt.show()

# 5. Age Distribution: Survived vs Not

plt.figure(figsize=(8, 5))
sns.histplot(data=df, x='Age', hue='Survived', bins=30, kde=True, palette='h
plt.title("Age Distribution by Survival")
plt.xlabel("Age")
plt.show()

# 6. Combine Sex & Class

plt.figure(figsize=(8, 6))
sns.catplot(x="Pclass", hue="Survived", col="Sex", data=df, kind="count", pa
plt.subplots_adjust(top=0.85)
plt.suptitle("Survival by Class & Gender")
plt.show()

# 7. Survival by Family (SibSp + Parch)

df['FamilySize'] = df['SibSp'] + df['Parch']
plt.figure(figsize=(6, 4))
 sns.barplot(x='FamilySize', y='Survived', data=df, palette='flare')
plt.title("Survival Rate by Family Size")
plt.ylabel("Survival Rate")
plt.show()

<ipython-input-19-cf2e0f97e5be>:6: FutureWarning:

Passing `palette` without assigning `hue` is deprecated and will be removed

in v0.14.0. Assign the `x` variable to `hue` and set `legend=False` for the
same effect.

sns.countplot(x='Survived', data=df, palette='Set2')

<ipython-input-19-cf2e0f97e5be>:13: FutureWarning:

Passing `palette` without assigning `hue` is deprecated and will be removed

in v0.14.0. Assign the `x` variable to `hue` and set `legend=False` for the
same effect.

sns.barplot(x='Sex', y='Survived', data=df, palette='pastel')

<ipython-input-19-cf2e0f97e5be>:20: FutureWarning:

Passing `palette` without assigning `hue` is deprecated and will be removed

in v0.14.0. Assign the `x` variable to `hue` and set `legend=False` for the
same effect.

sns.barplot(x='Pclass', y='Survived', data=df, palette='muted')

<ipython-input-19-cf2e0f97e5be>:27: FutureWarning:

Passing `palette` without assigning `hue` is deprecated and will be removed

in v0.14.0. Assign the `x` variable to `hue` and set `legend=False` for the
same effect.

sns.barplot(x='Embarked', y='Survived', data=df, palette='coolwarm')

<Figure size 800x600 with 0 Axes>

<ipython-input-19-cf2e0f97e5be>:49: FutureWarning:

Passing `palette` without assigning `hue` is deprecated and will be removed

in v0.14.0. Assign the `x` variable to `hue` and set `legend=False` for the
same effect.

sns.barplot(x='FamilySize', y='Survived', data=df, palette='flare')

Women had a survival rate of ~74%, compared to ~18% for men.

1st class passengers had ~63% survival rate, much higher than 3rd class
(~24%).
Children (age < 10) had higher survival rates, supporting the "women and
children first" approach.

Passengers who paid higher fares were more likely to survive.

Embarked from Cherbourg (C) had better outcomes (~55% survival).

Small families (1–2 siblings/spouses) showed higher survival rates than those
alone or in large groups.

Overall, being female, young, in a higher class, and having family nearby
improved chances of survival.

This notebook was converted with convert.ploomber.io