Statistical Analysis

import pandas as pd

# Load data
df = pd.read_excel('excel26deenero.xlsx')

# Calculate descriptive statistics
descriptive_stats = df[[
    'Likes', 
    'Comentarios', 
    'Visualizaciones', 
    'Duración del video (s)'
]].describe()

print("Estadísticas Descriptivas:")
print(descriptive_stats)

import seaborn as sns
import matplotlib.pyplot as plt

# Select numeric columns
numeric_cols = ['Likes', 'Comentarios', 'Visualizaciones', 'Duración del video (s)']

# Calculate correlation matrix
correlation_matrix = df[numeric_cols].corr()

print(correlation_matrix)

# Create heatmap visualization
fig, ax = plt.subplots(figsize=(10, 8))

sns.heatmap(
    df[numeric_cols].corr(), 
    annot=True, 
    cmap='coolwarm', 
    center=0,
    square=True,
    linewidths=1,
    cbar_kws={"shrink": 0.8},
    ax=ax
)

ax.set_title('Mapa de Calor de Correlaciones', fontsize=16, pad=20)
plt.tight_layout()
plt.show()

import numpy as np

# Analyze likes distribution
print(f"Mean Likes: {df['Likes'].mean():.2f}")
print(f"Median Likes: {df['Likes'].median():.2f}")
print(f"Mode: Median < Mean indicates right-skewed distribution")

# Visualize distribution
fig, ax = plt.subplots(figsize=(12, 6))
ax.hist(df['Likes'], bins=30, edgecolor='black', alpha=0.7)
ax.axvline(df['Likes'].mean(), color='red', linestyle='--', label='Mean')
ax.axvline(df['Likes'].median(), color='green', linestyle='--', label='Median')
ax.set_xlabel('Likes')
ax.set_ylabel('Frequency')
ax.set_title('Distribution of Likes across Videos')
ax.legend()
plt.show()

# Analyze comments distribution
print(f"Mean Comments: {df['Comentarios'].mean():.2f}")
print(f"Median Comments: {df['Comentarios'].median():.2f}")
print(f"Comments per Like Ratio: {df['Comentarios'].sum() / df['Likes'].sum():.4f}")

# Views analysis
print(f"Mean Views: {df['Visualizaciones'].mean():.2f}")
print(f"Median Views: {df['Visualizaciones'].median():.2f}")
print(f"View-to-Like Conversion: {(df['Likes'].sum() / df['Visualizaciones'].sum()) * 100:.2f}%")

# Group by category and calculate statistics
category_stats = df.groupby('Categoria').agg({
    'Likes': ['count', 'mean', 'sum'],
    'Comentarios': ['mean', 'sum'],
    'Visualizaciones': ['mean', 'sum']
}).round(2)

print(category_stats)

# Calculate coefficient of variation by category
cv_by_category = df.groupby('Categoria').agg({
    'Likes': lambda x: (x.std() / x.mean()) * 100,
    'Comentarios': lambda x: (x.std() / x.mean()) * 100
}).round(2)

print("Coefficient of Variation (%)")
print(cv_by_category)

# Analyze video duration
print(f"Average Duration: {df['Duración del video (s)'].mean():.2f} seconds")
print(f"Median Duration: {df['Duración del video (s)'].median():.2f} seconds")
print(f"Range: {df['Duración del video (s)'].min():.0f}s - {df['Duración del video (s)'].max():.0f}s")

# Correlation between duration and engagement
fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(15, 5))

# Duration vs Likes
ax1.scatter(df['Duración del video (s)'], df['Likes'], alpha=0.6)
ax1.set_xlabel('Duration (seconds)')
ax1.set_ylabel('Likes')
ax1.set_title('Duration vs Likes')

# Duration vs Views
ax2.scatter(df['Duración del video (s)'], df['Visualizaciones'], alpha=0.6, color='orange')
ax2.set_xlabel('Duration (seconds)')
ax2.set_ylabel('Views')
ax2.set_title('Duration vs Views')

plt.tight_layout()
plt.show()