# ===================================================================
# MODULE 10, LESSON 6: SUSTAINABLE BANKING & GREEN FINTECH
# ===================================================================

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
from datetime import datetime, timedelta
import warnings
warnings.filterwarnings('ignore')

print("="*70)
print("GREEN BANKING – ESG DATA PIPELINE & CARBON SCORING")
print("="*70)

# ----------------------------------------------------------------
# PART A: SIMULATING ESG DATA INGESTION (Multiple Sources)
# ----------------------------------------------------------------
print("\n" + "-"*60)
print("PART A: ESG Data Aggregation from Multiple Sources")
print("-"*60)

np.random.seed(42)

# Simulating corporate client ESG scores (Source 1: External ESG Rating Agency)
corporate_clients = pd.DataFrame({
    'client_id': [f'C{str(i).zfill(4)}' for i in range(1, 51)],
    'sector': np.random.choice(
        ['Energy', 'Manufacturing', 'Technology', 'Retail', 'Transportation', 'Real Estate', 'Financial Services'],
        50,
        p=[0.15, 0.2, 0.2, 0.15, 0.1, 0.1, 0.1]
    ),
    'revenue': np.random.uniform(10, 500, 50).round(2),  # $M
    'esg_score_third_party': np.random.randint(30, 95, 50),  # External agency rating (0-100)
    'carbon_intensity': np.random.uniform(0.1, 2.5, 50).round(2),  # tCO2e / $M revenue
    'green_revenue_pct': np.random.uniform(0, 80, 50).round(1)  # % of revenue from green products
})

# Simulating internal bank ESG data (Source 2: Internal Transactions & Operations)
corporate_clients['internal_carbon_estimate'] = corporate_clients['carbon_intensity'] * np.random.uniform(0.8, 1.2, 50)
corporate_clients['sustainability_risk'] = np.random.choice(['Low', 'Medium', 'High'], 50, p=[0.4, 0.4, 0.2])

print("Corporate Client ESG Dataset (Sample):")
print(corporate_clients.head(8).to_string(index=False))

# ----------------------------------------------------------------
# PART B: COMPOSITE ESG SCORE CALCULATION (Weighted Aggregation)
# ----------------------------------------------------------------
print("\n" + "-"*60)
print("PART B: Composite ESG Score & Green Classification")
print("-"*60)

# Weights for scoring
weights = {
    'esg_score_third_party': 0.35,
    'carbon_intensity': 0.25,  # Lower is better (inverted)
    'green_revenue_pct': 0.20,
    'sustainability_risk_mapping': 0.20
}

# Map sustainability risk to numeric
risk_mapping = {'Low': 90, 'Medium': 60, 'High': 30}
corporate_clients['risk_numeric'] = corporate_clients['sustainability_risk'].map(risk_mapping)

# Invert carbon intensity (higher carbon = lower score)
max_carbon = corporate_clients['carbon_intensity'].max()
corporate_clients['carbon_score'] = (1 - (corporate_clients['carbon_intensity'] / max_carbon)) * 100

# Calculate Composite ESG Score
corporate_clients['esg_composite_score'] = (
    weights['esg_score_third_party'] * corporate_clients['esg_score_third_party'] +
    weights['carbon_intensity'] * corporate_clients['carbon_score'] +
    weights['green_revenue_pct'] * corporate_clients['green_revenue_pct'] +
    weights['sustainability_risk_mapping'] * corporate_clients['risk_numeric']
).round(2)

# Classify clients into Green tiers
def classify_esg(score):
    if score >= 75:
        return 'Green (Low Carbon)'
    elif score >= 50:
        return 'Amber (Transitioning)'
    else:
        return 'Red (High Carbon)'

corporate_clients['esg_classification'] = corporate_clients['esg_composite_score'].apply(classify_esg)

print("\nComposite ESG Scores & Classification:")
print(corporate_clients[['client_id', 'sector', 'esg_composite_score', 'esg_classification']].head(10).to_string(index=False))

# ----------------------------------------------------------------
# PART C: PORTFOLIO CARBON FOOTPRINT DASHBOARD
# ----------------------------------------------------------------
print("\n" + "-"*60)
print("PART C: Portfolio Carbon Footprint Dashboard")
print("-"*60)

# Calculate portfolio-level metrics
portfolio_carbon = {
    'Total Clients': len(corporate_clients),
    'Average ESG Score': corporate_clients['esg_composite_score'].mean(),
    'Total Carbon Intensity (Portfolio)': corporate_clients['carbon_intensity'].sum(),
    'Average Carbon Intensity': corporate_clients['carbon_intensity'].mean(),
    'Green Clients (%)': (corporate_clients['esg_classification'] == 'Green (Low Carbon)').mean() * 100,
    'High Carbon Clients (%)': (corporate_clients['esg_classification'] == 'Red (High Carbon)').mean() * 100,
    'Total Green Revenue ($M)': corporate_clients['green_revenue_pct'].sum(),
}

print("\nPortfolio ESG Summary:")
for key, value in portfolio_carbon.items():
    if isinstance(value, float):
        print(f"  {key}: {value:.2f}")
    else:
        print(f"  {key}: {value}")

# ----------------------------------------------------------------
# PART D: VISUALIZATION – ESG SCORE DISTRIBUTION BY SECTOR
# ----------------------------------------------------------------
print("\n" + "-"*60)
print("PART D: ESG Performance by Sector (Visualization)")
print("-"*60)

# Create the visualization
fig, axes = plt.subplots(1, 2, figsize=(14, 6))

# Chart 1: ESG Score Distribution by Sector
sector_esg = corporate_clients.groupby('sector')['esg_composite_score'].agg(['mean', 'std']).reset_index()
sector_esg.columns = ['Sector', 'Mean_ESG', 'Std_ESG']
sector_esg = sector_esg.sort_values('Mean_ESG', ascending=False)

axes[0].barh(sector_esg['Sector'], sector_esg['Mean_ESG'], xerr=sector_esg['Std_ESG'], 
             color='green', alpha=0.7, edgecolor='black')
axes[0].axvline(x=75, color='red', linestyle='--', label='Green Threshold (75)')
axes[0].set_xlabel('Composite ESG Score')
axes[0].set_title('ESG Performance by Sector')
axes[0].legend()
axes[0].grid(True, alpha=0.3, axis='x')

# Chart 2: Client Classification Pie Chart
classification_counts = corporate_clients['esg_classification'].value_counts()
colors = {'Green (Low Carbon)': '#2ecc71', 'Amber (Transitioning)': '#f39c12', 'Red (High Carbon)': '#e74c3c'}
patches, texts, autotexts = axes[1].pie(
    classification_counts.values, 
    labels=classification_counts.index,
    autopct='%1.1f%%',
    colors=[colors.get(c, '#95a5a6') for c in classification_counts.index],
    startangle=90,
    explode=[0.05, 0, 0],
    shadow=True
)
axes[1].set_title('Portfolio ESG Classification')

plt.tight_layout()
plt.savefig('esg_portfolio_dashboard.png', dpi=300, bbox_inches='tight')
plt.show()
print("ESG Dashboard saved as 'esg_portfolio_dashboard.png'")

# ----------------------------------------------------------------
# PART E: GREEN LENDING DECISION ENGINE
# ----------------------------------------------------------------
print("\n" + "-"*60)
print("PART E: Green Lending Decision Engine")
print("-"*60)

class GreenLendingEngine:
    """
    Simulates an AI-driven decision engine for green loans.
    """
    def __init__(self, client_data):
        self.client_data = client_data
        
    def assess_loan_eligibility(self, client_id, loan_amount, loan_term_years):
        """Determines green loan eligibility and interest rate."""
        client = self.client_data[self.client_data['client_id'] == client_id]
        
        if client.empty:
            return {"error": "Client not found"}
        
        esg_score = client['esg_composite_score'].values[0]
        sector = client['sector'].values[0]
        carbon_intensity = client['carbon_intensity'].values[0]
        
        # Base interest rate
        base_rate = 8.5  # Standard bank rate
        
        # Green discount (lower rates for better ESG)
        if esg_score >= 75:
            discount = 2.0  # 2% discount
            decision = "Approved (Green Tier)"
        elif esg_score >= 50:
            discount = 0.75  # 0.75% discount
            decision = "Approved (Transitioning Tier)"
        else:
            discount = 0.0
            decision = "Conditional (Carbon Reduction Plan Required)"
            
        # Additional sector-specific adjustments
        if sector in ['Energy', 'Transportation'] and carbon_intensity > 1.5:
            decision += " - High carbon sector, additional scrutiny"
            discount = max(0, discount - 0.5)
            
        final_rate = base_rate - discount
        
        # Calculate repayment (simple amortization)
        monthly_rate = (final_rate / 100) / 12
        months = loan_term_years * 12
        if monthly_rate > 0:
            monthly_payment = loan_amount * (monthly_rate * (1 + monthly_rate) ** months) / ((1 + monthly_rate) ** months - 1)
        else:
            monthly_payment = loan_amount / months
            
        return {
            'client_id': client_id,
            'sector': sector,
            'esg_score': esg_score,
            'base_rate': base_rate,
            'discount': discount,
            'final_rate': round(final_rate, 2),
            'decision': decision,
            'loan_amount': loan_amount,
            'term_years': loan_term_years,
            'monthly_payment': round(monthly_payment, 2)
        }

# Instantiate the engine
lending_engine = GreenLendingEngine(corporate_clients)

# Test different clients
test_clients = ['C0001', 'C0020', 'C0045']
for client_id in test_clients:
    result = lending_engine.assess_loan_eligibility(client_id, loan_amount=1000000, loan_term_years=10)
    print(f"\nGreen Lending Assessment for {client_id}:")
    for key, value in result.items():
        if isinstance(value, float):
            print(f"  {key}: ${value:,.2f}" if 'payment' in key or 'amount' in key else f"  {key}: {value:.2f}")
        else:
            print(f"  {key}: {value}")

# ----------------------------------------------------------------
# SECTION 6: SUMMARY FOR THE DATA PRACTITIONER
# ----------------------------------------------------------------
print("\n" + "="*70)
print("LESSON 6 SUMMARY FOR THE DATA PRACTITIONER")
print("="*70)

print("""
1. Sustainable banking is driven by regulation (TCFD, EU Taxonomy), investor pressure, and customer demand.
2. Financed emissions (Scope 3) account for ~95% of a bank's carbon footprint – this is where data matters most.
3. Green Products: Green loans, Sustainability-Linked Loans (SLLs), Green bonds, Carbon trading.
4. Data Practitioner's role includes:
   - Aggregating ESG data from multiple sources (external ratings, internal metrics).
   - Calculating composite ESG scores and portfolio carbon footprints.
   - Building decision engines for green lending (as demonstrated).
   - Ensuring data quality for mandatory regulatory reporting.
5. Technology Enablers: AI for ESG scoring, IoT for real-time KPI monitoring, Blockchain for carbon credit traceability.
6. Action: Map your organization's data sources for ESG reporting and identify gaps in carbon intensity data.
""")

print("="*70)
print("END OF LESSON 6 – MODULE 10")
print("="*70)