Data Analysis with Python: A Jupyter Notebook Example

This post demonstrates how to include Jupyter notebook content in your blog. Here’s an example data analysis workflow.

Introduction

In this notebook, we’ll analyze some sample data using Python, pandas, and matplotlib.

Setup and Imports

import pandas as pd
import matplotlib.pyplot as plt
import numpy as np
import seaborn as sns

# Set up plotting style
plt.style.use('seaborn-v0_8')
sns.set_palette("husl")

Load and Explore Data

# Create sample data
np.random.seed(42)
data = {
    'date': pd.date_range('2023-01-01', periods=100, freq='D'),
    'sales': np.random.normal(1000, 200, 100),
    'visitors': np.random.normal(500, 100, 100),
    'category': np.random.choice(['A', 'B', 'C'], 100)
}

df = pd.DataFrame(data)
df['conversion_rate'] = df['sales'] / df['visitors']

print("Dataset shape:", df.shape)
df.head()

Output:

Dataset shape: (100, 5)
       date        sales    visitors category  conversion_rate
0 2023-01-01   996.714153   426.967649        A         2.334225
1 2023-01-02   881.730204   563.407932        C         1.565011
2 2023-01-03  1291.544472   695.455727        C         1.856892
3 2023-01-04   953.177555   423.586466        A         2.249803
4 2023-01-05   865.408363   544.883183        C         1.588313

Data Visualization

# Create subplots
fig, axes = plt.subplots(2, 2, figsize=(12, 10))

# Sales over time
axes[0,0].plot(df['date'], df['sales'], alpha=0.7)
axes[0,0].set_title('Sales Over Time')
axes[0,0].set_xlabel('Date')
axes[0,0].set_ylabel('Sales')

# Visitors by category
category_visitors = df.groupby('category')['visitors'].mean()
axes[0,1].bar(category_visitors.index, category_visitors.values)
axes[0,1].set_title('Average Visitors by Category')
axes[0,1].set_xlabel('Category')
axes[0,1].set_ylabel('Average Visitors')

# Conversion rate distribution
axes[1,0].hist(df['conversion_rate'], bins=20, alpha=0.7)
axes[1,0].set_title('Conversion Rate Distribution')
axes[1,0].set_xlabel('Conversion Rate')
axes[1,0].set_ylabel('Frequency')

# Correlation heatmap
correlation_data = df[['sales', 'visitors', 'conversion_rate']].corr()
im = axes[1,1].imshow(correlation_data, cmap='coolwarm', aspect='auto')
axes[1,1].set_title('Correlation Matrix')
axes[1,1].set_xticks(range(len(correlation_data.columns)))
axes[1,1].set_yticks(range(len(correlation_data.columns)))
axes[1,1].set_xticklabels(correlation_data.columns, rotation=45)
axes[1,1].set_yticklabels(correlation_data.columns)

plt.tight_layout()
plt.show()

Statistical Analysis

# Summary statistics
print("Summary Statistics:")
print(df[['sales', 'visitors', 'conversion_rate']].describe())

# Correlation analysis
print("\nCorrelation Matrix:")
print(correlation_data)

Output:

Summary Statistics:
              sales     visitors  conversion_rate
count    100.000000   100.000000       100.000000
mean     973.633870   499.313471         1.969167
std      217.355072    98.063616         0.567890
min      444.175957   274.052987         0.927181
25%      824.018703   434.077389         1.590983
50%      984.016525   499.359181         1.933964
75%     1138.651039   565.743736         2.318605
max     1564.007081   759.375021         3.847052

Correlation Matrix:
                   sales  visitors  conversion_rate
sales           1.000000  0.018739         0.894123
visitors        0.018739  1.000000        -0.748925
conversion_rate 0.894123 -0.748925         1.000000

Key Insights

From our analysis, we can observe:

1. Sales Trend: Sales show some variability over time with an average of ~974 units
2. Category Performance: All categories have similar visitor counts
3. Conversion Rate: There’s a strong positive correlation (0.89) between sales and conversion rate
4. Visitors Impact: Interestingly, there’s a negative correlation (-0.75) between visitors and conversion rate

Conclusion

This example demonstrates how to present Jupyter notebook content in a blog post format. The combination of code, outputs, and explanations makes for engaging technical content.

---

You can find the original notebook file in the repository or create your own following this structure.