An open-source, locally-run web application for statistical analysis and data visualization. GRA provides a no-code interface for regression analysis, Bayesian modeling, ANOVA, time series analysis, and more.

• Data Management: Upload, clean, and manage datasets (CSV, Excel, JSON)
• Statistical Analysis:
  • OLS Regression
  • Bayesian Regression
  • Bayesian Model Averaging (BMA)
  • ANOVA
  • VARX (Vector Autoregression with Exogenous Variables)
  • Structural Models (2SLS, 3SLS, SUR)
• Data Preparation:
  • Column management (rename, type conversion, ordering)
  • Data cleaning and normalization
  • Stationarity fixes for time series
  • Date format detection and conversion
• Visualization: Interactive plots and charts
• Session Management: Save and manage analysis sessions
• Paper Organization: Organize research papers and associated analyses

• Python: 3.9 or higher
• pip: Python package manager
• R (optional): Required for Bayesian Model Averaging (BMA) analysis. If not installed, BMA features will be unavailable.

git clone https://github.com/pooryaselkghafari/Generative-Research-Assistant.git
cd Generative-Research-Assistant

python3 -m venv venv

On macOS/Linux:

source venv/bin/activate

On Windows:

venv\Scripts\activate

pip install -r requirements.txt

Note: If you encounter any dependency conflicts, you may need to upgrade pip first:

pip install --upgrade pip

python manage.py migrate

This will create the SQLite database file (db.sqlite3) and set up all necessary tables.

python manage.py collectstatic --noinput

python manage.py runserver

The application will be available at http://127.0.0.1:8000/

If port 8000 is already in use, specify a different port:

python manage.py runserver 8080

1. Navigate to the main page at http://127.0.0.1:8000/
2. Click on "Upload Dataset" or use the dataset upload interface
3. Select a file (CSV, Excel, or JSON format)
4. The dataset will be processed and added to your dataset list

1. Click on a dataset to open the data cleaner
2. In the cleaner interface, you can:
   • Manage Columns: Rename columns, change data types (numeric, categorical, ordinal, date, etc.)
   • Code Columns: Apply coding transformations
   • Normalize Columns: Apply normalization (min-max, mean centering, standardization)
   • Drop Columns: Remove unwanted columns
   • Fix Date Formats: Detect and convert date formats
3. Click "Save" to overwrite the original dataset, or "Save as..." to create a new dataset

1. From the main page, click "New Analysis" or select an existing session
2. Choose your analysis type:
   • OLS Regression: Standard linear regression
   • Bayesian Regression: Bayesian linear regression
   • BMA: Bayesian Model Averaging (requires R)
   • ANOVA: Analysis of Variance
   • VARX: Vector Autoregression with Exogenous Variables
   • Structural Models: 2SLS, 3SLS, SUR
3. Select your dependent and independent variables
4. Configure analysis options
5. Click "Run Analysis"

• Results are displayed with:
  • Statistical tables and coefficients
  • Model diagnostics
  • Interactive visualizations
  • Export options (CSV, Excel, JSON)

• Save Sessions: Analysis sessions are automatically saved
• Organize with Papers: Group related analyses into research papers
• Export Results: Download results in various formats
• Session History: View and manage previous analysis sessions

For VARX analysis:

• Stationarity Check: View ADF test results for each variable
• Fix Stationarity: Apply differencing or log transformations to make variables stationary
• IRF Plots: Generate Impulse Response Function plots
• Correlation Heatmaps: Visualize variable correlations

GRA/
├── engine/              # Main application code
│   ├── dataprep/       # Data preparation and cleaning
│   ├── models/         # Django database models
│   ├── services/       # Business logic services
│   ├── static/         # Static files (CSS, JS)
│   ├── templates/      # HTML templates
│   └── views/          # View controllers
├── models/             # Statistical analysis models
│   ├── regression.py
│   ├── bayesian_regression.py
│   ├── BMA.py
│   ├── ANOVA.py
│   ├── VARX.py
│   └── structural_model.py
├── data_prep/          # Data preparation utilities
├── statbox/            # Django project settings
├── manage.py           # Django management script
└── requirements.txt    # Python dependencies

The application runs locally by default with:

• Database: SQLite (no configuration needed)
• Authentication: Not required (local use only)
• Static Files: Served by Django development server

You can customize settings using environment variables:

export DEBUG=True
export SECRET_KEY=your-secret-key
export ALLOWED_HOSTS=localhost,127.0.0.1

If you encounter import errors, ensure your virtual environment is activated and all dependencies are installed:

pip install -r requirements.txt

If you encounter database errors, try resetting the database:

rm db.sqlite3
python manage.py migrate

If BMA analysis fails, ensure R is installed and accessible from your PATH:

# Check if R is installed
R --version

# Install required R packages (if needed)
Rscript -e "install.packages('BAS', repos='https://cloud.r-project.org')"

If port 8000 is already in use, specify a different port:

python manage.py runserver 8080

If static files (CSS, JS) are not loading:

python manage.py collectstatic --noinput

Ensure you're using Python 3.9 or higher:

python3 --version

If you have multiple Python versions, use:

python3.9 -m venv venv

python manage.py test

If you modify models:

python manage.py makemigrations
python manage.py migrate

The admin interface is available at /admin/ (no authentication required for local use).

This application runs entirely locally on your machine:

• All data is stored locally in your SQLite database
• No data is transmitted to external servers
• No tracking or analytics
• No cloud services
• You have full control over your data

This project is open source and available under the MIT License. See LICENSE file for details.

Contributions are welcome! Please feel free to submit a Pull Request.

For issues and questions, please open an issue on the GitHub repository.

This tool is built using:

• Django (Python web framework)
• Pandas (Data manipulation)
• NumPy (Numerical computing)
• Statsmodels (Statistical modeling)
• R (for Bayesian Model Averaging)
• Various visualization libraries

Note: This is a local application designed to run on your machine. No authentication or external services are required for basic usage.