Guyana Development Analysis: A Multi-Dimensional Perspective

A comprehensive data science project analyzing Guyana's development across economic, social, and environmental dimensions spanning 100+ years.

Overview

This project uses machine learning, satellite imagery analysis, and data visualization to tell the compelling story of Guyana's development from the 1920s to present day. It covers 12 key dimensions:

Economic Indicators

• 🏦 GDP & wealth trends (100 years)
• 📦 Export economy evolution (50 years)
• ⛏️ Gold & diamond production
• 🛢️ Oil production (2019-present)

Social Development

• 😊 Happiness index
• 📊 Economic inequality (Gini coefficient)
• 📚 Literacy rates and education

Environmental Analysis

• 🌿 Biodiversity metrics
• 🌳 Forest cover & deforestation
• 💧 Water resources dynamics

Features

• Time Series Forecasting: ARIMA, Prophet, VAR, LSTM models for economic predictions
• Clustering Analysis: K-means, DBSCAN for pattern recognition
• Satellite Imagery: Google Earth Engine integration for forest and water analysis
• Computer Vision: U-Net CNN for deforestation detection
• Interactive Visualizations: Plotly, Folium maps
• Correlation & Causal Analysis: Multi-dimensional relationship exploration

Installation

Prerequisites

• Python 3.9.6 or higher
• Google Earth Engine account (sign up at earthengine.google.com)

Setup

1. Clone the repository:

git clone <your-repo-url>
cd guyana-development-analysis

2. Create and activate virtual environment:

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

3. Install core dependencies:

pip install -r requirements.txt

4. (Optional) Install geospatial packages:

First install GDAL:

# On macOS
brew install gdal

# On Ubuntu/Debian
sudo apt-get install gdal-bin libgdal-dev

Then install geospatial requirements:

pip install -r requirements-geo.txt

5. Authenticate with Google Earth Engine:

earthengine authenticate

Project Structure

guyana-development-analysis/
│
├── README.md                          # This file
├── requirements.txt                   # Core Python dependencies
├── requirements-geo.txt               # Geospatial dependencies
├── .gitignore                         # Git ignore rules
│
├── guyana_analysis.ipynb              # MAIN COMPREHENSIVE NOTEBOOK
│
├── data/                              # Data directory (gitignored)
│   ├── raw/                           # Original downloaded data
│   │   ├── economic/                  # GDP, exports, wealth
│   │   ├── social/                    # Happiness, literacy
│   │   ├── resources/                 # Gold, diamond, oil
│   │   ├── environmental/             # Biodiversity
│   │   └── satellite/                 # GEE downloads, rasters
│   ├── processed/                     # Cleaned data
│   └── metadata/                      # Data source documentation
│
├── outputs/                           # Generated artifacts (gitignored)
│   ├── figures/                       # Static plots (PNG, SVG)
│   ├── interactive/                   # Interactive HTML plots
│   ├── models/                        # Saved ML models
│   └── reports/                       # Summary statistics
│
└── src/                               # Helper modules
    ├── data_fetchers.py              # API/data download functions
    ├── preprocessing.py               # Data cleaning utilities
    ├── satellite_utils.py             # GEE and raster processing
    └── visualization.py               # Reusable plot functions

Data Sources

All data comes from free, publicly available sources:

Economic Data

• World Bank Open Data: GDP, exports, poverty indicators
• UN Comtrade: Export composition by commodity
• IMF: Historical economic estimates

Resource Production

• USGS Minerals Yearbook: Gold & diamond production
• U.S. EIA: Oil production statistics
• World Bank Commodity Prices: Price data for context

Social Indicators

• World Happiness Report: Happiness scores and components
• UNESCO: Literacy rates and education statistics
• World Bank Poverty & Equity: Gini, income distribution

Environmental Data

• GBIF: Species occurrence records
• IUCN Red List: Threatened species
• Living Planet Index: Population trends

Satellite Imagery (Google Earth Engine)

• Hansen Global Forest Change: Tree cover (2000-2023)
• MODIS: NDVI vegetation indices
• JRC Global Surface Water: Water occurrence (1984-2024)
• Sentinel-2: High-resolution imagery

Usage

1. Start Jupyter Notebook:

jupyter notebook

2. Open guyana_analysis.ipynb

3. Run cells sequentially (first run will download data - may take time)

4. Outputs will be saved to outputs/ directory

Machine Learning Techniques

The notebook implements:

• Time Series: ARIMA, SARIMAX, Prophet, VAR, LSTM
• Clustering: K-means, DBSCAN, Hierarchical
• Classification: Random Forest, XGBoost
• Regression: Random Forest, GWR (Geographically Weighted)
• Computer Vision: U-Net for satellite image segmentation
• Anomaly Detection: Isolation Forest, LSTM Autoencoder
• Causal Inference: Difference-in-Differences, Granger Causality

Expected Outputs

Figures (~50 plots)

• GDP trajectories and forecasts
• Export composition changes
• Production trends
• Deforestation risk maps
• Biodiversity hotspots
• Correlation heatmaps

Interactive Visualizations (~15)

• GDP forecast with confidence intervals
• Animated GDP vs happiness scatter
• Forest change maps
• Flood risk maps

ML Models (~10)

• ARIMA (GDP forecasting)
• Prophet (exports)
• Random Forest (deforestation risk, happiness drivers)
• XGBoost (flood risk)
• U-Net weights (forest change detection)

Development Timeline

• Week 1: Project setup ✓
• Weeks 2-3: Data acquisition
• Week 4: Data preprocessing
• Week 5: Exploratory data analysis
• Weeks 6-7: ML analysis
• Week 8: Integrated analysis
• Week 9: Narrative & polish
• Week 10: Documentation & testing

Contributing

This is a personal/research project. If you'd like to contribute or have suggestions, please open an issue.

License

MIT License - See LICENSE file for details

Citation

If you use this analysis in your research, please cite:

Guyana Development Analysis: A Multi-Dimensional Perspective (2026)
Repository: <your-repo-url>

Acknowledgments

• Data Providers: World Bank, UN, IMF, NASA, ESA, USGS, UNESCO, GBIF, IUCN
• Tools: Python, Jupyter, scikit-learn, TensorFlow, Google Earth Engine, Plotly
• Inspiration: Showcasing Guyana's rich heritage and development journey

Contact

For questions or collaboration opportunities, please open an issue on GitHub.

---

Note: This project requires significant computational resources for satellite imagery analysis. Consider using cloud platforms (Google Colab, Kaggle) for GPU-accelerated tasks.