Real-Time Economic Indicators & Policy Insight Platform
EconoPulse is an interactive web dashboard for exploring economic indicators (inflation, GDP growth, unemployment, FX rates), comparing countries side-by-side, detecting economic stress signals, and forecasting trends.
| Feature | Description |
|---|---|
| 📊 Dashboard | KPI cards + interactive time-series charts for all indicators |
| 🌍 Country Comparison | Side-by-side line charts, bar charts, and correlation heatmaps |
| 🚨 Signal Detection | Automatic flagging of inflation spikes, GDP contractions, unemployment surges |
| 🔮 Forecasting | Holt linear-trend model (falls back to OLS) with configurable horizon |
| 💾 Export | Download filtered data and alerts as CSV, or a full text report |
| 🗂 Data Sources | Built-in sample data, World Bank Open Data API, or your own CSV upload |
Ten advanced analytical features are available in the ⭐ Premium tab:
| # | Feature | Description |
|---|---|---|
| 1 | 📊 Economic Health Score | Composite 0–100 index per country per year. GDP growth and GDP per capita raise the score; inflation and unemployment lower it (weighted, min-max normalised). |
| 2 | 🔮 Multi-Country Forecast Comparison | Forecast the same indicator for all selected countries simultaneously on a single chart using Holt's linear-trend model. |
| 3 | 🌡️ YoY Change Heatmap | Pivot heatmap of year-over-year absolute changes (percentage points) for any indicator, making acceleration and deceleration immediately visible. |
| 4 | 🔥 Economic Risk Score Heatmap | Annual risk score per country = sum of alert severity weights (low=1, medium=2, high=3). Heatmap surface highlights where and when stress concentrated. |
| 5 | 📉 Recession Detection | Identifies consecutive years of negative GDP growth (configurable minimum, default 2) and overlays shaded bands on the GDP chart. |
| 6 | 🏆 Country Percentile Rankings | Rank selected countries against each other for any indicator and year; bar chart coloured green-to-red by percentile. |
| 7 | 📈 Moving Average Trend Overlay | Configurable rolling-mean window (2–10 years) overlaid on the raw series to smooth noise and reveal medium-term trends. |
| 8 | 🎛️ Custom Alert Thresholds | Slider-controlled per-indicator thresholds replace the built-in defaults; alerts fire in real time as thresholds are adjusted. |
| 9 | 📊 Volatility Analysis | Rolling standard deviation of any indicator for each country; bar chart of average volatility ranks the most erratic economies. |
| 10 | 🔵 Economic Similarity Clustering | K-means clustering (k configurable) groups countries by their full indicator profile in a selected year; scatter plot and cluster-assignment table. |
Econopulse/
├── app.py # Main Streamlit application
├── requirements.txt # Python dependencies
├── data/
│ ├── fetcher.py # Data loading (World Bank API / CSV / sample)
│ └── sample_data.py # Offline sample data generator
├── analytics/
│ ├── signals.py # Economic stress signal detection
│ ├── forecasting.py # Time-series forecasting
│ └── premium.py # 10 premium analytical functions
├── utils/
│ └── export.py # CSV & text-report export helpers
└── tests/ # Pytest test suite (80 tests)
The easiest way to share EconoPulse publicly at no cost.
What you need:
- A free Streamlit Community Cloud account (sign in with GitHub)
- This repository pushed to a public GitHub repo (or a private repo on a paid Streamlit plan)
- No API keys — the World Bank API and the built-in sample data are both public
Steps:
- Go to https://share.streamlit.io and click "New app"
- Select your GitHub repository and branch
- Set Main file path to
app.py - Click Deploy — Streamlit installs
requirements.txtautomatically
Your app will be live at https://<your-handle>-econopulse-app-<hash>.streamlit.app.
What you need:
- Python 3.10 or later
pip
# 1. Clone the repository
git clone https://github.com/RayBen445/Econopulse.git
cd Econopulse
# 2. (Optional) create a virtual environment
python -m venv .venv
source .venv/bin/activate # Windows: .venv\Scripts\activate
# 3. Install dependencies
pip install -r requirements.txt
# 4. Launch the app
streamlit run app.pyOpen http://localhost:8501 in your browser.
What you need:
- Docker installed
Create a Dockerfile in the project root:
FROM python:3.12-slim
WORKDIR /app
COPY requirements.txt .
RUN pip install --no-cache-dir -r requirements.txt
COPY . .
EXPOSE 8501
CMD ["streamlit", "run", "app.py", "--server.port=8501", "--server.address=0.0.0.0"]Then build and run:
docker build -t econopulse .
docker run -p 8501:8501 econopulseOpen http://localhost:8501.
This Docker image can be pushed to any container registry (Docker Hub, AWS ECR, Google Artifact Registry, Azure Container Registry) and deployed on:
- AWS — Elastic Container Service (ECS) / App Runner
- Google Cloud — Cloud Run (
gcloud run deploy) - Azure — Container Apps / App Service
- Railway / Render / Fly.io — point to your Docker image or GitHub repo
What you need:
- Heroku CLI and a free/paid Heroku account
Create a Procfile in the project root:
web: streamlit run app.py --server.port=$PORT --server.address=0.0.0.0
Then deploy:
heroku create econopulse-app
git push heroku main
heroku open| Item | Required? | Notes |
|---|---|---|
| Python 3.10+ | ✅ Yes | All platforms except Streamlit Cloud (managed) |
requirements.txt dependencies |
✅ Yes | Installed automatically on Streamlit Cloud |
| GitHub repository | ✅ For Streamlit Cloud | Public repo for the free tier |
| World Bank API key | ❌ No | The API is open and does not require authentication |
| Database / storage | ❌ No | All data is fetched at runtime or uploaded by the user |
| Environment variables / secrets | ❌ No | No secrets needed for any current feature |
| Custom CSV data | ⚙️ Optional | Upload via the sidebar "Upload CSV" option at runtime |
pip install pytest
pytest tests/ -v- Python 3.10+
- Streamlit — UI framework
- Pandas / NumPy — data manipulation
- Plotly — interactive charts
- statsmodels — Holt's linear-trend forecasting
- scikit-learn — linear regression fallback
- World Bank Open Data API — live economic data (optional)
This project is licensed under the MIT License.