📊 Sri Lanka Fertility Rate Forecasting - ML Project

A machine learning project that predicts Sri Lanka's adolescent fertility rates using 64 years of World Bank data (1960-2023). The project achieves 78% accuracy in forecasting and reveals a remarkable 78.7% decline in fertility rates over six decades.

🎯 Quick Start

# Clone the repository
git clone <your-repo-url>
cd fertility-forecasting-sri-lanka

# Set up environment
python3 -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt

# Run the analysis
python fertility_analysis_minimal.py

📈 Key Results

• Best Model: Linear Regression (MAE: 3.326)
• Historical Trend: 78.7% decline from 1960 to 2023
• 2030 Prediction: 4.4 births per 1,000 women
• Policy Success: World-class fertility reduction achievement

📁 Project Structure

├── README.md                           # This file
├── PROJECT_REPORT.md                   # Detailed analysis report
├── requirements.txt                    # Python dependencies
├── .gitignore                         # Git ignore rules
├── datasets/
│   └── social-development_lka.csv     # World Bank data
├── fertility_analysis_minimal.py      # Main analysis script
├── fertility_forecasting.py          # Comprehensive framework
├── fertility_analysis_simple.py      # Extended analysis with plots
└── generated_files/
    ├── PROJECT_SUMMARY.md
    ├── dataset_analysis.py
    └── forecasting_project_proposal.md

🚀 Reproduction Instructions

Step 1: Environment Setup

Prerequisites:

• Python 3.8 or higher
• pip package manager
• 2GB free disk space

Create Virtual Environment:

# Create virtual environment
python3 -m venv venv

# Activate it
source venv/bin/activate  # Linux/Mac
# OR
venv\Scripts\activate     # Windows

Step 2: Install Dependencies

# Install required packages
pip install pandas numpy matplotlib seaborn scikit-learn

# Optional: For advanced models
pip install prophet tensorflow statsmodels plotly dash

Or use requirements file:

pip install -r requirements.txt

Step 3: Run Analysis

Quick Analysis (5 minutes):

python fertility_analysis_minimal.py

Full Analysis with Visualizations (15 minutes):

python fertility_analysis_simple.py

Comprehensive Framework (30 minutes):

python fertility_forecasting.py

Step 4: View Results

The analysis will output:

• Model performance metrics
• Future predictions (2024-2030)
• Key insights and trends
• Generated visualizations (if applicable)

📊 Expected Output

Sri Lanka Adolescent Fertility Rate Analysis
==================================================
Loading data...
Data loaded: 64 observations from 1960 to 2023

Basic Statistics:
Mean: 32.64
Std: 13.44
Range: 15.1 to 71.0

Trend Analysis:
1960: 71.0 births per 1,000 women
2023: 15.1 births per 1,000 women
Total decline: -78.7%

Forecasting Models:
Linear     - MAE: 3.326, RMSE: 3.561
Quadratic  - MAE: 9.522, RMSE: 9.687
Cubic      - MAE: 8.340, RMSE: 9.852

Best model: Linear

Future Forecasts:
2024: 8.75
2025: 8.03
2026: 7.31
2027: 6.59
2028: 5.87
2029: 5.15
2030: 4.43

==================================================
KEY INSIGHTS
==================================================
• Remarkable 79% decline over 63 years
• Consistent downward trend
• Current rate among lowest in South Asia
• Projected 2030 rate: 4.4
• Policy success story in family planning

Analysis completed successfully!

🔧 Requirements

System Requirements

• OS: Windows 10+, macOS 10.14+, or Linux
• Python: 3.8, 3.9, 3.10, or 3.11
• RAM: 4GB minimum (8GB recommended)
• Storage: 2GB free space

Python Dependencies

pandas>=1.3.0
numpy>=1.21.0
scikit-learn>=1.0.0
matplotlib>=3.4.0
seaborn>=0.11.0

Optional (for advanced features):

prophet>=1.1.0
tensorflow>=2.8.0
statsmodels>=0.13.0
plotly>=5.0.0
dash>=2.0.0

📈 Dataset Information

Source: World Bank Social Development Indicators Country: Sri Lanka (LKA) Indicator: Adolescent fertility rate (births per 1,000 women ages 15-19) Code: SP.ADO.TFRT Time Range: 1960-2023 Observations: 64 years of complete data Missing Values: None

File: datasets/social-development_lka.csv Size: ~800KB (845 total indicators, 32 social development metrics)

🧪 Testing Your Setup

Run this quick test to verify everything works:

# test_setup.py
import pandas as pd
import numpy as np
from sklearn.linear_model import LinearRegression
import matplotlib.pyplot as plt

print("✅ All packages imported successfully!")
print(f"Python version: {python.version}")
print(f"Pandas version: {pd.__version__}")
print(f"NumPy version: {np.__version__}")
print(f"Scikit-learn version: {sklearn.__version__}")

🔍 Troubleshooting

Common Issues

Issue 1: Module not found error

# Solution: Install missing packages
pip install <missing-package-name>

Issue 2: Permission errors (Windows)

# Solution: Run as administrator or use --user flag
pip install --user <package-name>

Issue 3: Python version compatibility

# Solution: Check Python version
python --version
# Upgrade if needed (Python 3.8+ required)

Issue 4: Data file not found

FileNotFoundError: datasets/social-development_lka.csv

# Solution: Ensure you're in the correct directory
ls datasets/  # Should show the CSV file

Getting Help

1. Check Error Messages: Read the full error traceback
2. Verify File Paths: Ensure all files are in correct locations
3. Test Environment: Run the test script above
4. Check Dependencies: Verify all packages are installed
5. Python Version: Ensure you're using Python 3.8+

📖 Usage Examples

Basic Prediction

import pandas as pd
from sklearn.linear_model import LinearRegression
import numpy as np

# Load data
data = pd.read_csv("datasets/social-development_lka.csv")
fertility_data = data[data['Indicator Code'] == 'SP.ADO.TFRT'].copy()

# Prepare features
X = fertility_data['Year'].values.reshape(-1, 1)
y = fertility_data['Value'].values

# Train model
model = LinearRegression()
model.fit(X, y)

# Predict future
future_year = np.array([[2025]])
prediction = model.predict(future_year)
print(f"2025 prediction: {prediction[0]:.2f}")

Custom Year Range

# Predict custom year range
start_year = 2024
end_year = 2035
future_years = np.array(range(start_year, end_year + 1)).reshape(-1, 1)
predictions = model.predict(future_years)

for year, pred in zip(range(start_year, end_year + 1), predictions):
    print(f"{year}: {pred:.2f}")

🎓 Educational Use

This project is designed for:

• Machine Learning Students: Learn regression, model evaluation, time series
• Data Science Courses: Practical application of ML to real-world data
• Public Policy Analysis: Understanding social development trends
• Research Projects: Reproducible analysis template

Learning Objectives

1. Data Preprocessing: Handle real-world CSV data
2. Regression Modeling: Compare different algorithms
3. Model Evaluation: Use MAE, RMSE metrics appropriately
4. Time Series: Understand temporal data challenges
5. Business Insights: Translate technical results to actionable recommendations

🤝 Contributing

Want to improve this project? Here's how:

1. Fork the repository
2. Create feature branch: git checkout -b feature/your-feature
3. Make changes: Improve code, documentation, or analysis
4. Test thoroughly: Ensure everything works
5. Submit pull request: Describe your improvements

Contribution Ideas:

• Add more sophisticated models (ARIMA, Neural Networks)
• Create interactive dashboards
• Add confidence intervals
• Compare with other countries
• Improve visualizations

📄 License

This project is open source and available under the MIT License.

📞 Support

Having issues?

• Check the troubleshooting section above
• Review error messages carefully
• Verify all requirements are met
• Test with the provided examples

For technical questions:

• Open an issue on GitHub
• Include your error message and system info
• Describe steps to reproduce the problem

🏆 Acknowledgments

• Data Source: World Bank Open Data
• Sri Lanka: Remarkable policy achievement over 60+ years
• Open Source Community: Python, scikit-learn, pandas developers
• Educational Purpose: Advancing ML education through real-world examples

---

Ready to explore Sri Lanka's remarkable fertility decline story through data science? Start with the Quick Start section above!