ettomarett/Wealth-And-Rainfall-Analysis-Morocco

# Morocco Wealth-Rainfall Analysis Dashboard ## Project Overview This project explores the relationship between wealth and rainfall patterns across Morocco's regions, incorporating advanced forecasting models for future rainfall predictions. It integrates geospatial, economic, and meteorological data to provide interactive visualizations and data-driven insights for researchers, policymakers, and the public. ## Key Findings 1. **Rainfall Forecasting**: - Prophet model shows highest accuracy (MAE: 8.4131) - Holt-Winters demonstrates strong performance (MAE: 8.9054) - Modern approaches outperform traditional methods - Seasonal patterns significantly impact prediction accuracy 2. **Wealth-Rainfall Relationship**: - Analysis of regional wealth distribution - Correlation studies between rainfall patterns and economic indicators - Impact assessment of rainfall variability on different regions - Identification of vulnerable areas ## Features - Unified dataset combining wealth and rainfall data at the regional level - Interactive Streamlit dashboard with: - Rainfall and wealth data exploration - Geographic visualizations (choropleth maps, wealth points) - Advanced forecasting models - Correlation and seasonal analysis - Data dictionary and methodology explanations - Efficient data processing pipeline - Machine learning models for rainfall prediction ## Data Sources - **Rainfall Data:** Historical rainfall measurements for Moroccan regions - **Wealth Data:** Relative Wealth Index (RWI) with geospatial coordinates - **Administrative Boundaries:** GeoJSON and CSV files for Morocco's regions ## Environment Setup ### 1. Create and Activate Conda Environment ```bash conda create -n dddm_env python=3.10 conda activate dddm_env ``` ### 2. Install Requirements ```bash pip install -r requirements.txt ``` ### 3. Prepare Data - Place raw datasets in the `Datasets/` directory - Run the data processing script: ```bash python process_data.py ``` ### 4. Launch the Dashboard ```b …