Pandas Data Analysis & Visualization

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
from scipy import stats
from datetime import datetime, timedelta

class DataAnalyzer:
    def __init__(self, data_path: str):
        self.df = pd.read_csv(data_path)
        self.setup_analysis()
    
    def setup_analysis(self):
        """Initial data setup and cleaning"""
        # Handle missing values
        self.df = self.df.dropna(subset=['revenue', 'customer_id'])
        
        # Convert date columns
        self.df['date'] = pd.to_datetime(self.df['date'])
        
        # Create derived features
        self.df['month'] = self.df['date'].dt.month
        self.df['quarter'] = self.df['date'].dt.quarter
        self.df['revenue_per_customer'] = self.df['revenue'] / self.df['customers']
    
    def generate_insights(self):
        """Generate comprehensive business insights"""
        insights = {}
        
        # Revenue analysis
        insights['total_revenue'] = self.df['revenue'].sum()
        insights['avg_monthly_revenue'] = self.df.groupby('month')['revenue'].mean()
        insights['revenue_growth'] = self.calculate_growth_rate('revenue')
        
        # Customer analysis
        insights['customer_retention'] = self.calculate_retention_rate()
        insights['top_customers'] = self.df.nlargest(10, 'revenue_per_customer')
        
        # Statistical analysis
        insights['revenue_correlation'] = self.df[['revenue', 'customers', 'marketing_spend']].corr()
        
        return insights
    
    def calculate_growth_rate(self, column: str) -> float:
        """Calculate month-over-month growth rate"""
        monthly_data = self.df.groupby('month')[column].sum()
        growth_rates = monthly_data.pct_change().dropna()
        return growth_rates.mean() * 100
    
    def calculate_retention_rate(self) -> float:
        """Calculate customer retention rate"""
        monthly_customers = self.df.groupby('month')['customer_id'].nunique()
        retention_rates = []
        
        for i in range(1, len(monthly_customers)):
            current_customers = set(self.df[self.df['month'] == monthly_customers.index[i]]['customer_id'])
            previous_customers = set(self.df[self.df['month'] == monthly_customers.index[i-1]]['customer_id'])
            
            retained = len(current_customers.intersection(previous_customers))
            retention_rate = retained / len(previous_customers) if previous_customers else 0
            retention_rates.append(retention_rate)
        
        return np.mean(retention_rates) * 100
    
    def create_dashboard_plots(self):
        """Create comprehensive visualization dashboard"""
        fig, axes = plt.subplots(2, 2, figsize=(15, 12))
        
        # Revenue trend
        monthly_revenue = self.df.groupby('month')['revenue'].sum()
        axes[0, 0].plot(monthly_revenue.index, monthly_revenue.values, marker='o')
        axes[0, 0].set_title('Monthly Revenue Trend')
        axes[0, 0].set_xlabel('Month')
        axes[0, 0].set_ylabel('Revenue ($)')
        
        # Customer distribution
        sns.histplot(data=self.df, x='revenue_per_customer', bins=30, ax=axes[0, 1])
        axes[0, 1].set_title('Revenue per Customer Distribution')
        
        # Correlation heatmap
        correlation_matrix = self.df[['revenue', 'customers', 'marketing_spend']].corr()
        sns.heatmap(correlation_matrix, annot=True, cmap='coolwarm', ax=axes[1, 0])
        axes[1, 0].set_title('Feature Correlation Matrix')
        
        # Quarterly performance
        quarterly_data = self.df.groupby('quarter')['revenue'].sum()
        axes[1, 1].bar(quarterly_data.index, quarterly_data.values)
        axes[1, 1].set_title('Quarterly Revenue Performance')
        axes[1, 1].set_xlabel('Quarter')
        axes[1, 1].set_ylabel('Revenue ($)')
        
        plt.tight_layout()
        return fig