Nakrevive/Rwanda_food_prices-2000-2025-and-predictions

Food prices in Rwanda have been rising significantly, affecting affordability and food security. This project analyzes food price trends, volatility, and cost of a standard food basket using official data. # Rwanda_food_prices-2000-2025-and-predictions Food prices in Rwanda have been rising significantly, affecting affordability and food security. This project analyzes food price trends, volatility, and cost of a standard food basket using official data. Dataset - File used: `Rwanda_food_prices_cleaned.csv` - Date Range: 2000–2025 - Rows: ~80,000+ - Markets:20+ Rwandan markets - Commodities: Rice, beans, maize, meat, vegetables, fruits, cooking oil, etc. -Source: World Food Programme (WFP) – Rwanda Market Price Monitoring This cleaned dataset includes: - Standardized commodity names - Converted units into KG - Parsed and cleaned dates - Price per kg (`price_per_kg`) - Monthly groupings - Total food basket cost Project Objectives 1. Analyze the price trends of major food commodities in Rwanda. 2. Compare food affordability between markets (urban vs rural). 3. Build a Rwandan Food Basket and track inflation over time. 4. Detect price shocks and high-volatility periods. 5. Visualize findings using Matplotlib, Seaborn, and Tableau. 6. Provide insights that policymakers or humanitarian agencies can use. Technologies Used - Python (Pandas, NumPy, Matplotlib, Seaborn) - Jupyter Notebook - Excel - SQL - Tableau - GitHub Data Cleaning Steps The raw dataset contained inconsistent formats, mixed units, missing values, and noise. I performed: 1. Date Cleaning - Parsed multiple inconsistent date formats - Ensured chronological alignment - Extracted `month_start` for monthly analysis 2. Commodity Standardization - Merged duplicates (e.g., “rice”, “Rice”, “Rice (imported)”) - Unified all commodities into a consistent naming system 3. Unit Conversion - Converted all measurements (grams, kilograms, litres, pieces) into **kilograms** - Created a universal column: `price_per_kg` 4. Market Cleaning - Standardized market names - Removed invalid entries 5. Missing Value Treatment - Dropped or imputed values depending on severity 6. Monthly Aggregation - Calculated average price …