Data Science Analysis Project

Data Loading & Initial Exploration

Load the dataset and perform initial exploration:

• Import required libraries (numpy, pandas, matplotlib)
• Load CSV using pd.read_csv()
• Display shape, dtypes, and first/last 5 rows
• Use df.info() and df.describe()
• Check for missing values with df.isnull().sum()

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

# Load data
df = pd.read_csv('sales_data.csv')

# Initial exploration
print(f"Dataset Shape: {df.shape}")
print(f"\nData Types:\n{df.dtypes}")
print(f"\nMissing Values:\n{df.isnull().sum()}")

Data Cleaning

Clean the dataset by handling all data quality issues:

• Handle missing values (fill or drop with justification)
• Remove or fix invalid data (negative quantities)
• Convert order_date to datetime type
• Check for and handle duplicates
• Create a new column: total_amount = quantity * unit_price

# Handle missing values
df['unit_price'].fillna(df['unit_price'].median(), inplace=True)
df['region'].fillna('Unknown', inplace=True)

# Remove invalid quantities
df = df[df['quantity'] > 0]

# Convert date
df['order_date'] = pd.to_datetime(df['order_date'])

# Create total_amount column
df['total_amount'] = df['quantity'] * df['unit_price']

NumPy Statistical Analysis

Use NumPy to compute statistics on numeric columns:

• Calculate mean, median, std using NumPy functions
• Find min, max, percentiles (25th, 50th, 75th)
• Create a NumPy array from the total_amount column
• Use boolean indexing to filter high-value orders (> $100)
• Calculate correlation between quantity and total_amount

# NumPy statistics
amounts = df['total_amount'].values

print(f"Mean: ${np.mean(amounts):.2f}")
print(f"Median: ${np.median(amounts):.2f}")
print(f"Std Dev: ${np.std(amounts):.2f}")
print(f"25th Percentile: ${np.percentile(amounts, 25):.2f}")
print(f"75th Percentile: ${np.percentile(amounts, 75):.2f}")

# High-value orders
high_value = amounts[amounts > 100]
print(f"\nHigh-value orders (>$100): {len(high_value)}")

Pandas GroupBy Analysis

Perform aggregations using Pandas groupby:

• Total sales by category
• Average order value by region
• Top 5 customers by total spending
• Sales count by payment method
• Best-selling products by quantity

# Sales by category
category_sales = df.groupby('category')['total_amount'].sum().sort_values(ascending=False)

# Average order by region
region_avg = df.groupby('region')['total_amount'].mean()

# Top 5 customers
top_customers = df.groupby('customer_id')['total_amount'].sum().nlargest(5)

Time Series Analysis

Analyze sales trends over time:

• Extract day of week, day name from order_date
• Calculate daily total sales
• Find the busiest day of the week
• Calculate rolling 3-day average sales

Data Visualizations (Minimum 6)

Create at least 6 professional visualizations using Matplotlib:

• Bar Chart: Sales by category
• Pie Chart: Sales distribution by region
• Line Plot: Daily sales trend
• Histogram: Distribution of order values
• Horizontal Bar: Top 5 products by revenue
• Grouped Bar: Sales by category per region

# Example: Sales by category bar chart
fig, ax = plt.subplots(figsize=(10, 6))
category_sales.plot(kind='bar', color=['#6366f1', '#10b981', '#f59e0b', '#ef4444'], ax=ax)
ax.set_title('Total Sales by Category', fontsize=14, fontweight='bold')
ax.set_xlabel('Category')
ax.set_ylabel('Total Sales ($)')
ax.tick_params(axis='x', rotation=45)
plt.tight_layout()
plt.savefig('charts/sales_by_category.png', dpi=300)
plt.show()

Business Insights Summary

Write a markdown summary with key findings:

• What category generates the most revenue?
• Which region has the highest average order value?
• Who are the top 3 customers and what do they buy?
• What day of the week has the most sales?
• 3 actionable recommendations for the business

Export Results

Save your analysis outputs:

• Export cleaned DataFrame to cleaned_sales_data.csv
• Save all charts to a charts/ folder
• Create a summary statistics CSV file